Install the package remotes:
install.packages("remotes")
Install the package prospect with the following command line in R session:
remotes::install_github('jbferet/prospect')
prosail uses Support Vector Regression (SVR) for hybrid inversion.
The default SVM implementation is currently based on the package
liquidSVM.
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 will 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 liquidSVM is not properly installed, prosail automatically switches to
the ksvm function of the kernlab
package with nu regression and RBF kernel.
This is also accessible if setting method <- 'nu-svr' when calling
train_prosail_inversion and prosail_hybrid_train.
An SVM implementation based on the R package caret
is also available, with linear (method <- 'nu-svr') or RBF
(method <- 'nu-svr') kernel.
**WARNING : ** optimal performances are obtained with liquidSVM. caret implementation may need significantly longer time for training and application stages.
The package prosail can then be installed with the following command line in R session:
remotes::install_github('jbferet/prosail')
**WARNING : ** many functions have been renamed in v3.0 of the package. The documentation has been updated accordingly. Please refer to the documentation for additional information.
The tutorial vignettes start 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 (Institute 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