This title appears in the Scientific Report :
2023
Please use the identifier:
http://dx.doi.org/10.1109/IGARSS52108.2023.10281579 in citations.
Fast Machine Learning Simulator of At-Sensor Radiances for Solar-Induced Fluorescence Retrieval with DESIS and Hyplant
Fast Machine Learning Simulator of At-Sensor Radiances for Solar-Induced Fluorescence Retrieval with DESIS and Hyplant
In many remote sensing applications the measured radi-ance needs to be corrected for atmospheric effects to studysurface properties such as reflectance, temperature or emis-sion features. The correction often applies radiative transferto simulate atmospheric propagation, a time-consuming stepusually...
Saved in:
Personal Name(s): | Pato, Miguel |
---|---|
Alonso, Kevin / Auer, Stefan / Buffat, Jim / Carmona, Emiliano / Maier, Stefan / Müller, Rupert / Rademske, Patrick / Rascher, Uwe / Scharr, Hanno | |
Contributing Institute: |
Datenanalyse und Maschinenlernen; IAS-8 |
Imprint: |
IEEE
2023
|
Physical Description: |
7563-7566 |
DOI: |
10.1109/IGARSS52108.2023.10281579 |
Conference: | IGARSS 2023 - 2023 IEEE International Geoscience and Remote Sensing Symposium, Pasadena (CA), 2023-07-16 - 2023-07-21 |
Document Type: |
Contribution to a conference proceedings |
Research Program: |
Cross-Domain Algorithms, Tools, Methods Labs (ATMLs) and Research Groups |
Publikationsportal JuSER |
In many remote sensing applications the measured radi-ance needs to be corrected for atmospheric effects to studysurface properties such as reflectance, temperature or emis-sion features. The correction often applies radiative transferto simulate atmospheric propagation, a time-consuming stepusually done offline. In principle, an efficient machine learn-ing (ML) model can accelerate the simulation step. This is thegoal pursued here in the context of solar-induced fluorescence(SIF) emitted by vegetation around the O2-A band using thespaceborne DESIS and airborne HyPlant spectrometers. Wepresent an ML simulator of at-sensor radiances trained onsynthetic spectra and describe its performance in detail. Thesimulator is fast and accurate, constituting a promising alter-native to a full-fledged, lengthy radiative transfer code for SIFretrieval in the O2-A band with DESIS and HyPlant.Index Terms— solar-induced fluorescence, hyperspectralsensors, radiative transfer, machine learning |