devtools
Install the package devtools
following the instructions depending on your operating system.
prospect
Install the package prospect
with the following command line in R session:
devtools::install_github('jbferet/prospect')
prosail
uses Support Vector Regression (SVR) for hybrid inversion.
The default SVM implementation is currently based on the package
liquidSVM
.
A SVM implementation based on the R package caret
is also available.
liquidSVM
provides very efficient and
fully-integrated hyper-parameter selection, multithreading and GPU support.
However, this package is not maintained anymore and may cause difficulties during
the installation.
To install liquidSVM
, please follow installation instructions provided in the
documentation webpage.
Once liquidSVM
is installed, you may need to add the 32bit DLL into the R library.
This i386
directory should be downloaded here
and copied into the local directory on your computer, where the binary codes of liquidSVM are installed:
Path_For_My_R_distribution/library/liquidSVM/libs/
.
liquidSVM
is a suggested package, so the installation of prosail
should succeed
even without liquidSVM
.
Two main functions using liquidSVM
as default may be impacted: train_prosail_inversion
and PROSAIL_Hybrid_Train
.
If the installation of liquidSVM
does not succeed, please define method = 'svmRadial'
or method = 'svmLinear'
as optional input variable.
This will use the caret implementation, but it will require longer training stage.
prosail
The package prosail
can then be installed with the following command line in R session:
devtools::install_github('jbferet/prosail')
A tutorial vignette is available here.
This research was supported by the Agence Nationale de la Recherche (ANR, France) through the young researchers project BioCop (ANR-17-CE32-0001)
We thank Ingo Steinwart and Philipp Thomann (nstitute for Stochastics and Applications, University of Stuttgart, Germany) for the development of the package liquidSVM
.
If you use prosail, please consider citing the following references when appropriate :
Féret, J.-B. & de Boissieu, F. (2024). prospect
: an R package to link leaf optical properties with their chemical and structural properties with the leaf model PROSPECT. Journal of Open Source Software, 9(94), 6027, https://doi.org/10.21105/joss.06027
Féret J-B, Gitelson AA, Noble SD & Jacquemoud S, 2017. PROSPECT-D: Towards modeling leaf optical properties through a complete lifecycle. Remote Sensing of Environment, 193, 204–215. https://doi.org/10.1016/j.rse.2017.03.004
Féret J-B, Berger K, de Boissieu F & Malenovský Z, 2021. PROSPECT-PRO for estimating content of nitrogen-containing leaf proteins and other carbon-based constituents. Remote Sensing of Environment, 252, 112173. https://doi.org/10.1016/j.rse.2020.112173
Jacquemoud S, Verhoef W, Baret F, Bacour C, Zarco-Tejada PJ, Asner GP, François C & Ustin SL, 2009. PROSPECT+ SAIL models: A review of use for vegetation characterization. Remote Sensing of Environment, 113:S56–S66. https://doi.org/doi:10.1016/j.rse.2008.01.026
Berger K, Atzberger C, Danner M, D’Urso G, Mauser W, Vuolo F & Hank T 2018. Evaluation of the PROSAIL Model Capabilities for Future Hyperspectral Model Environments: A Review Study. Remote Sensing, 10:85. https://doi.org/10.3390/rs10010085
Verhoef W & Bach H, 2007. Coupled soil–leaf-canopy and atmosphere radiative transfer modeling to simulate hyperspectral multi-angular surface reflectance and TOA radiance data. Remote Sensing of Environment, 109:166-182. https://doi.org/10.1016/j.rse.2006.12.013
Verhoef W, Jia L, Xiao Q & Su Z, 2007. Unified optical-thermal four-stream radiative transfer theory for homogeneous vegetation canopies. IEEE Transactions in Geosciences and Remote Sensing, 45:1808–1822. https://doi.org/10.1109/TGRS.2007.895844
Steinwart I & Thomann P (2017). liquidSVM: A Fast and Versatile SVM package. ArXiv e-prints 1702.06899, http://www.isa.uni-stuttgart.de/software
Jacquemoud S, Baret F, Hanocq J-F, 1992. Modeling spectral and bidirectional soil reflectance. Remote Sensing of Environment, 41, 123–132. https://doi.org/10.1016/0034-4257(92)90072-R