The Atmospheric Ghost Correction

TL;DR

This paper introduces a spectral characterization of atmospheric ghosts, providing a framework to improve atmospheric correction algorithms and quantify ghosting effects in remote sensing data.

Contribution

It presents the first spectral analysis of ghost types and integrates these properties into atmospheric correction, advancing remote sensing accuracy.

Findings

Spectral properties of three ghost types characterized

Improved atmospheric correction framework demonstrated

Preliminary analysis of ghosting variations over space and time

Abstract

While the presence of ghosts has been known for decades, the impact of these apparitions on remote sensing observations has gone unquantified, leaving atmospheric corrections susceptible to ghosting. In this work, we present the first spectral characterization of three common ghost types and provide a framework for incorporating these properties into atmospheric correction algorithms. We demonstrate the effect of this improved atmospheric ghost correction compared to an atmosphere-only implementation. Lastly, we discuss some preliminary spatiotemporal variations in haunting intensity and type.