Conference proceeding
Temporal Mapping of Hyperspectral Data
2019 10th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS), Vol.2019-, pp.1-4
09/2019
DOI: 10.1109/WHISPERS.2019.8921373
Abstract
The increasing popularity of hyperspectral sensors is dramatically increasing the temporal availability of data. To date, algorithms struggle to compare hyperspectral data collected across dates due to different environmental conditions during collection. In this work, we develop a temporal mapping in order to map data collected from one year to a different year. We investigated both conditional generative adversarial networks (cGANs) as well as affine transformations to perform this mapping. Both methods showed an improvement over using data from past collections without mapping, with cGANs outperforming the affine transformation.
Details
- Title: Subtitle
- Temporal Mapping of Hyperspectral Data
- Creators
- Ronald Fick - University of FloridaPaul Gader - University of FloridaAlina Zare - University of FloridaSusan Meerdink - University of Florida
- Resource Type
- Conference proceeding
- Publication Details
- 2019 10th Workshop on Hyperspectral Imaging and Signal Processing: Evolution in Remote Sensing (WHISPERS), Vol.2019-, pp.1-4
- Publisher
- IEEE
- DOI
- 10.1109/WHISPERS.2019.8921373
- ISSN
- 2158-6276
- eISSN
- 2158-6276
- Language
- English
- Date published
- 09/2019
- Academic Unit
- Geographical and Sustainability Sciences
- Record Identifier
- 9984259632202771
Metrics
3 Record Views