Name
Title | ||
|---|---|---|
Comparison of radiative transfer in physics-based models for an improved understanding of empirical hyperspectral data |
Abstract | ||
|---|---|---|
This paper examines the methodology of detecting targets in airborne or satellite hyperspectral imagery using physics based models. More specifically, the radiative transfer inherently coupled to various physical models is considered. In fact, taking into account atmospheric effects is crucial in target detection applications, especially when dealing with targets that are particularly difficult to detect. Many tools have been developed independently which incorporate physical models that simulate atmospheric radiation transfer. Some (e.g. DIRSIG) predict sensor-reaching radiance while others (e.g. FLAASH, ATREM) retrieve ground-leaving reflectance by removing atmospheric effects. With the final aim of performing forward modeling target detection on a particularly challenging scenario, this paper illustrates the preliminary study carried out in order to assess the physical model employed and achieve a better data understanding before proceeding to detection. A cross-comparison between some well-known and established models, in addition to forward modeling, is examined. Results reveal the need for better understanding of real data by identifying the major sources of uncertainty. The strong impact of atmospheric condition uncertainty and adjacency effects, along with, though to a lesser extent, further inaccuracy introduced by possible calibration and spectral library measurement errors, are all factors that will be investigated in future work. |
Year | DOI | Venue |
|---|---|---|
2009 | 10.1109/WHISPERS.2009.5288986 | WHISPERS |
Keywords | Field | DocType |
geophysical signal processing,object detection,physics,remote sensing,spectral analysis,airborne hyperspectral imagery,atmospheric effects,empirical hyperspectral data,forward modeling,physics-based models,radiative transfer,satellite hyperspectral imagery,target detection,dirsig,flaash,physics-based model,data models,reflectivity,atmospheric modeling,physical model,measurement error,artificial neural networks,hyperspectral sensors,mathematical model,hyperspectral imagery | Data modeling,Object detection,Physical model,Remote sensing,Hyperspectral imaging,Atmospheric model,Radiative transfer,Radiance,Observational error | Conference |
ISBN | Citations | PageRank |
978-1-4244-4687-2 | 2 | 0.52 |
References | Authors | |
2 | 3 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Stefania Matteoli | 1 | 152 | 18.05 |
| Emmett J. Ientilucci | 2 | 20 | 4.02 |
| John P. Kerekes | 3 | 194 | 35.38 |