Log in Register

Log in

Welcome to the ARTMO website!

 

--- Please register to freely download ARTMO and get access to the manuals and tutorials. By registering you agree with the License agreement*. You will first receive an email to verify your email address (if nothing receiving, check Spam!). Click on the link. We will then verify whether you are a non-commercial user and activate your account. If successful you will be informed by email.  For some countries we noticed registration problems; if not receiving an email, please contact us.The Download page will appear after logging in.---

 

Plant Radiative Transfer models (RTMs) have become important tools for the analysis of optical Earth observation data, providing meaningful links between radiometry and environmental applications, such as Earth system, ecological processes and precision agriculture. However, for the broader community these models are still often perceived as excessively complicated tools and are not always easily accessible.

The in-house developed Automated Radiative Transfer Models Operator (ARTMO) Graphic User Interface (GUI) is a software package that provides essential tools for running and inverting a suite of plant RTMs, both at the leaf and at the canopy level. ARTMO facilitates consistent and intuitive user interaction, thereby streamlining model setup, running, storing and spectra output plotting for any kind of optical sensor operating in the visible, near-infrared and shortwave infrared range (400-2500 nm). 

ARTMO also hosts the Atmospheric Look-up Generator (ALG) and the Decomposition and Analysis of Time Series software (DATimeS) as standalone software packages. 

 

                
ARTMO, ALG & DATimeS.

Principally, ARTMO allows:

  1. To configure and run leaf and canopy RTMs, independently or combined, in an intuitive way through various GUIs with input options to insert single values, value ranges, or imported external datasets.
  2. To simulate and store a massive quantity of spectral output based on a look-up table (LUT) approach in a relational database.
  3. To plot groups of simulated spectra in the same plotting window with color gradients as a function of input parameters.
  4. To export simulated spectra and associated meta-data to a text file for further processing.
  5. To analyze and apply retrieval techniques in order generate maps of biophysical parameters from optical remote sensing imagery.
 

ARTMO’s v.3.31  hierarchical design with all its modules as of August 2023.

 

 

* License agreement for ARTMO, its models, tools and toolboxes.

The programs contained in this package are granted free of charge for research and education purposes only. This thus excludes commercial activities, including its use for acquiring or executing National or International projects (e.g. EU, ESA, US). For commercial users a license file with purpose ´Commercial´ can be requested.  Scientific results produced using the software provided shall acknowledge the use of this implementation provided by us (see Publications).

If you plan to use it for non-scientific purposes, don't hesitate to contact us. Because the programs are licensed free of charge, there is no warranty for the program, to the extent permitted by applicable law. except when otherwise stated in writing the copyright holders and/or other parties provide the program "as is" without warranty of any kind, either expressed or implied, including, but not limited to, the implied warranties of merchantability and fitness for a particular purpose. The entire risk as to the quality and  performance of the program is with you. Should the program prove defective, you assume the cost of all necessary servicing, repair or correction. In no event unless required by applicable law or agreed to in writing will any copyright holder, or any other party who may modify and/or redistribute the program, be liable to you for damages, including any general, special, incidental or consequential damages arising out of the use or inability to use the program (including but not limited to loss of data or data being rendered inaccurate or losses sustained by you or third parties or a failure of the program to operate with any other programs), even if such holder or other party has been advised of the possibility of such damages.

  

 

Novelties:

This table provides an overivew of ARTMO's novelties starting from 23 July 2014.

Date 

Novelties
23/07/2014
  • ARTMO essentials has been updated to 3.04 in order to be compatible with the implementation of new models: the leaf model LIBERTY and the combined model SCOPE. Also a bug when functioning without Sensor module has been corrected.
  • The leaf model LIBERTY has been implemented. A Matlab version of LIBERTY (v1.0) has been developed by Yang Xi-guang.
  • SCOPE v1.53 has been integrated into ARTMO (also referred to as A-SCOPE). SCOPE has been expanded with various biochemical models during the course of ESA's 'FLEX/Sentinel 3 Mission Photosynthesis Study'. 
17/01/2015
  • ARTMO essentials has been updated to v. 3.10. Several new features have been implemented. Most importantly, it is now possible to generate fully random LUTs. Further, some bugs have been resolved.
  • The forest canopy model INFORM (Atzberger & Schlerf) has been implemented (v. 1.00). INFORM can be coupled with leaf optical models.
  • The SI and MLRA retrieval toolboxes have been updated to v. 1.10. A cross-validation sub-sampling module has been added to these toolboxes.
  • The processing and storing of maps in the the MLRA toolbox has been synchronized with that of the SI toolbox.
  • Regarding the SI toolbox, it is now also possible to evaluate the performance of single bands. An option to plot a triangular correlation matrix has been added.
12/02/2015
  • A bug has been corrected in the MLRA toolbox. No new version number is given. 
09/03/2015
  • A bug has been identified when adding a new Sensor. We have resolved this in ARTMO essentials. It has been upgraded to v. 3.11.
21/03/2015
  • A bug was identified and resolved regarding the visualization of LUT boundaries and number of samples. It led to ARTMO essentials v. 3.12.
  • The SI toolbox has been updated to v. 1.11. Several small utilities have been corrected: 1:1-plotting, statistics, loading of User data and displaying correlation matrix as a triangle. 
  • The LUT-based Inversion toolbox has been updated to v. 101. Several bugs have been corrected; using derivative, 1:1-spectral plotting, displaying correlation matrix. 
26/03/2015
  • The MLRA toolbox has been updated to v. 1.11.Several small bugs  have been identied and corrected: 1:1-plotting, R2-adj error handling.
03/04/2015
  • An option has been added to control the generation of random numbers (Seed). ARTMO essentials has been upgraded to v. 3.13.
23/04/2015
  • ARTMO essential has been updated to v. 3.14. A bug with Sensor and coupled RTMs for PROSPECT-5 and LIBERTY has been corrected.
  • The MLRA and Sprectral Indices toolbox has been updated to v.12. Now more flexibility is allowed when reading text files with User data. 
  • A spectral resampling module has been added (v. 1.00).
16/06/2015
  • ARTMO essential has been updated to v. 3.15. It is now possible to export large LUTs to a text file.
  • Some minor bugs in Sensor (v. 1.03) and Spectral resample (v. 1.01) tools have been resolved.
  • The three retrieval toolboxes have been upgraded (SI: v. 1.13, MLRA: v. 1.13 and LUT-inversion: v. 1.03) with various new functions. Most importantly, it is now possible to process TIFF files and out-of-memory problems in case of big images have been resolved. Also the option to optimize and process retrieval strategies per land cover class has been revised and improved. 
26/06/2015
  • A Global Sensitivity Analysis (GSA) toolbox has been added (v. 1.00). 
27/07/2015
  • An Emulator toolbox has ben added (v1.02). 
  • ARTMO has been updated to v. 3.16 with minor updates.
  • The three retrieval toolboxes have been upgraded (SI: v. 1.14, MLRA: v. 1.14 and LUT-inversion: v. 1.04) with improved functions. Most importantly, the work flow has been made more logic and some bugs were corrected.  

09/11/2015
  • ARTMO essential has been updated to v. 3.17. Most importanly, it is now possible to store simulations directly into NetCDF files.
  • The Sensor module has been updated with new default sensor settings, e.g. Landsat-8 (v. 1.04).
  • The three Retrieval toolboxes have been updated with new functions and a log window (SI: v. 1.15, MLRA: v. 1.15 and LUT-inversion: v. 1.05). For instance, now not only images but also .txt files (e.g. coming from field spectrometers) can be processed.
  • The Emulator toolbox has been updated with a few more pseudo MO-MLRAs and a log window (v. 1.03).
  • The GSA toolbox underwent some cosmetic changes (v. 1.01).
28/03/2016
  • ARTMO essential has been updated to v. 3.18 with minor updates. For instance, a Latin Hypercube Sampling has been added.
  • The Emulator toolbox has been updated with various new options to inspect emulation accuracies (v. 1.04). 
  • The three Retrieval toolboxes have been updated with small post-processing tools (SI: v. 1.16, MLRA: v. 1.16 and LUT-inversion: v. 1.06). For instance,for the SI and MLRA toolboxes the portability of developed models can be inspected by validating against external data. The LUT-inversion toolbox can now also invert an external LUT.
  • The GSA toolbox can now also analyze models as developed by the Emulator toolbox.  (v. 1.02).
  • ARTMO is in a process to become Linux-proof. However, there may still some MySQL issues to be resolved. 
04/04/2016
  • A small bug was encountered in the Emulator toolbox.  This has been fixed. No new version number is given. 
  • A lost 'keyboard' (breakpoint) remained in the GSA tool. This has been fixed. No new version number is given. 
06/06/2016
  • There was a bug encountered in the previous version in case of systematically combining PROSPECT with SAIL simulations. This has been resolved (ARTMO v. 3.19).
  • The SI toolbox has been significantly improved (v. 1.17). Results are now much faster stored into a MySQL DB. Also, an option is added to inspect results for a selected band combination. 
15/06/2016
  • A bug was encountered in the GSA toolbox. This has been corrected now (v. 1.03).  However, parallel computing leads to an error in newer Matlab versions (v. 2014 onwards). This can be resolved by setting # cores to 1 (see manual). 
04/08/2016
  • A few minor bugs were corrected in the GSA toolbox (v. 1.04). 
18/10/2016
  • ARTMO essential has been updated to v. 3.20 with a few new RTMs: the leaf model Fluspect-B (Vilfan et al., 2016), and to PROSPECT-4 a COSINE extension (Jay et al., 2016). 
  • The Emulator toolbox has been updated with a new option to inspect emulation accuracies per wavelength (v. 1.05). Also some bugs have been corrected.
  • The three Retrieval toolboxes have been updated with small post-processing tools and some bugs were corrected (SI: v. 1.18, MLRA: v. 1.17 and LUT-inversion: v. 1.07). For instance, to the MLRA toolbox an active learning (AL) module has been added.
  • The GSA toolbox has been made compatible with Fluspect-B and COSINE to PROSPECT-4.  (v. 1.05).
09/02/2017
  • ARTMO essential has been updated to v. 3.21. In this version various bugs are fixed and a license is introduced so that commercial users can also try ARTMO. 
  • The three Retrieval toolboxes have been updated (SI: v. 1.19, MLRA: v. 1.18 and LUT-inversion: v. 1.08). Some bugs have been corrected and the processing of images has been made more user-friendly.  
  • The Emulator toolbox has been updated with some more machine learning algorithms and made more user-friendly (v. 1.06).

13/02/2017
  • We have  updated ARTMO essentials with some fixes related to the generation of the license file. No new version number is given. 

17/02/2017
  • We have  updated ARTMO essentials with some fixes related to the generation of the license file and also fixed some bugs in LUT-inversion. No new version number is given. 
28/02/2017
  • We have  updated ARTMO essentials with some fixes related to the generation of the license file. No new version number is given (v. 3.21.3). 
28/03/2017
  • We have  updated ARTMO essentials (v. 3.21.4) and the LUT-inversion toolbox (v.108.2) with some fixes regarding inversion of .txt file. 
19/07/2017
  • ARTMO essentials has been updated to v. 3.22. In this version PROSPECT-D is introduced and SCOPE is updated to v. 1.70. Various small bugs have been corrected.
  • The three Retrieval toolboxes have been updated (SI: v. 1.20, MLRA: v. 1.19 and LUT-inversion: v. 1.09). The cross-validation module has been improved. Now Leave-One-Out works properly. The dimensionality reduction module in MLRA has been improved. Some bugs have been corrected. 
  • The Emulator toolbox has been updated with GPR now running more computationally efficient (v. 1.07).
28/07/2017
  • A few remaining minor bugs have been corrected in LUT inversion (v. 1.09.1), SI (v. 1.20.1) and in ARTMO essentials (v. 3.22.1). No new version number has been given.
27/12/2017
  • ARTMO essentials has been updated to v. 3.23. Some small bugs have been corrected. Database meta-info has been added. 
  • The three Retrieval toolboxes have been updated (SI: v. 1.21, MLRA: v. 1.20 and LUT-inversion: v. 1.10).  Some small utilities have been added and bugs corrected. 
  • The Emulator toolbox has been improved with new utilities and LUT and Txt emulator running faster (v. 1.08).
  • The Spectral resampling (v. 1.02)  and Sensor (v. 1.05) tools have been improved. 
  • A new Automated Scene Generator Module (A-SGM) tool has been introduced to generate scenes (v. 1.12).
12/01/2018
  • A remaining bug has been corrected in MLRA toolbox (v. 1.20.1). No new version number has been given.
30/07/2018
  • ARTMO essentials has been updated to v. 3.24. Multiple small bugs have been corrected and various GUIs have been made more user-friendly. Some bugs in the running of SCOPE v. 1.70 has been corrected.
  • The three Retrieval toolboxes have been updated (SI: v. 1.22, MLRA: v. 1.21 and LUT-inversion: v. 1.11).  Some small utilities have been added and bugs were corrected. New machine learning algorithms have been added in the MLRA toolbox. An improved spectral noise model has been added.
  • The Emulator toolbox has been expanded with a scene generation module and new machine learning algorithms (v. 1.08).
  • The Spectral resampling (v. 1.03)  and Sensor (v. 1.06) tools have been improved with new utilities. 
  • The A-SGM tool to generate scenes has been improved (v. 1.13).
  • The GSA tool has been updated with an utility to apply GSA to spectral indices. (v. 1.07).

03/09/2018
  • The MLRA toolbox has been updated to v 1.21.1. A bug in a MLRA algorithm has been corrected. In the toolbox no new version number has been given.
26/12/2018
  • ARTMO essentials has been updated to v. 3.25. Multiple small bugs have been corrected and various GUIs have been made more user-friendly. Several sampling distribution functions have been added. Some bugs in the running of SCOPE v. 1.70 has been corrected.
  • The three Retrieval toolboxes have been updated (SI: v. 1.23, MLRA: v. 1.22 and LUT-inversion: v. 1.12).  Some small utilities have been added and bugs were corrected.
  • The Emulator toolbox (v. 1.09), the GSA (v. 1.08), Spectral resampling (v. 1.04)  and Sensor (v. 1.07) tools have been improved with new utilities. 
  • ALG v. 2.0 has been released. ALG is a standalone software package for running amtospheric RTMs (6S, MODTRAN).

08/08/2019

 ARTMO:
  •  ARTMO essentials has been updated to v.3.26. A new leaf model has been added “PROSPECT-DyN”. Some bugs have been fixed and small improvements added, e.g. adding of soil spectra for SAIL and INFORM.
  • A few new tools have been added: (1) TOC2TOA (v. 1.00) that enables coupling with ALG-generated atmospheric LUTs with ARTMO LUTs; (2) BRDFplot (v.1.00) that enables plotting polar plots; and (3) LabelMe (v.1.00) that enables generating a LUT based on images and associated maps.
  • The three Retrieval toolboxes have been updated (SI: v.1.24, MLRA: v.1.23 and LUT-inversion: v.1.13).  Some small utilities have been added and a few bugs have been corrected.
  • Some minor bugs in the Emulator toolbox (v.1.11) and the Global Sensitivity Analysis toolbox (v.1.09) have been corrected and small improvements added.
  • The Spectral resampling (v. 1.05) and Sensor (v. 1.08) tools have been improved with new utilities, e.g. filling in NaN values and resampling sensors. 

ALG (v2.0):

  • New models are now compatible with ALG: MODTRAN6 and libRadtran (in mode Disort)
  • An Aerosol Toolkit (v 1.0.0) has been included to store user-defined aerosol properties. It currently works with 6SV, MODTRAN5 and MODTRAN6.
  • ALG now allows to insert user-defined atmospheric profiles in LUTs for the models MODTRAN5 and MODTRAN6.
  • Users can re-use the distribution of LUT nodes from a previously generated LUT.
  • A new MODTRAN interrogation technique was included in ALG based on the work of Verhoef and Bach 2003.
  • MODTRAN transmission mode has been included in ALG. Users can now make LUTs of gaseous transmittance.
 05/01/2020

ARTMO:

  • ARTMO essentials has been updated to v.3.27. A new leaf model has been added: “FluspectB-CX”. Some bugs have been fixed and small improvements added, e.g. saving output as NetCDF.
  • The three Retrieval toolboxes have been updated (SI: v.1.25, MLRA: v.1.24 and LUT-inversion: v.1.14).  Some small utilities have been added and a few bugs have been corrected. For instance output maps can be saved to NetCDF.
  • The Spectral resampling (v. 1.06) and Sensor (v. 1.08) tools have been improved with new utilities, e.g. removing (noisy) spectral regions and fill it up with an interpolation method. 
  • Some improvements have been implemented into the TOC2TOA toolbox (v. 1.01) and into the LabelMe toolbox (v. 1.01). For instance, the TOC2TOA now also provides TOA reflectance output.

ALG (v2,1):

  • Users can now run libRadtran in the “gas transmission” mode, based on the direct transmission output of libRadtran with the two-stream solver.
  • An Atmosphere Generator Toolkit (v 1.0.0) has been included to store user-defined atmospheric profiles. It currently works with 6SV, MODTRAN5, MODTRAN6 and libRadtran.
  • A Plotting tool (v 1.0.0) has been included in ALG v2.1 to visualize various input and output files related with ALG (e.g. LUT, OPAC files, 6SV Mie files, user-defined aerosol database).
  • Some minor bugs were corrected.
 08/01/2020

We have identified and fixed a remaining bug in ARTMO essentials v. 3.27. It has been reuploaded to the Download page (no new version number was given). 
13/01/2020

Some users have identified a few more bugs in ARTMO essentials v. 3.27.  They have been fixed  and ARTMO has been reuploaded to the Download page (no new version number was given). 
15/05/2020

We have released the first version of th DATimeS (v1.10). DATimeS is a time series toolbox for reconstruction of images and phenology indicators calculation. 
 30/12/2020

ARTMO:

  • ARTMO essentials has been updated to v.3.28. A new leaf model has been added: “PROSPECT-PRO” (Feret al., 2020). Some bugs have been fixed and for the leaf models now also absorptance can be generated.
  • The SI and MLRA Retrieval toolboxes have been updated (SI: v.1.26, MLRA: v.1.25).  Some bugs have been fixed and small utilities have been added. For instance several more machine learning algorithms have been implemented and the active learning module has been improved.
  • Some improvements have been implemented into the GSA toolbox (v.1.01), Emulator toolbox (v.1.13), LabelMe tool (v.1.01) and Sensor tool (v.1.10). For instance, the emulator toolbox allows now emulation of hyperspectral-to-hyperspectral data.

DATimeS (v1.10) as plugin of ARTMO:

  • Generating spatially continuous maps from discontinuous data using advanced machine learning fitting algorithms,
  • Detecting heterogeneous spatial patterns of phenological indicators (i.e., crop key growth stages) throughout multiple seasons,
  • Synergy of multiple time series data sources through multi-output Gaussian processes regression.

ALG (v3.0):

  •  ALG’s graphical user interface has been renewed and adapted to the latest Matlab R2020b App Designer tool.
  • A TOA radiance mode (only for Lambertian surface) is now available with MODTRAN, 6SV and libRadtran.
  • ALG is now compatible with the polradtran solver in libRadtran.
  • For Windows users, the compatibility of libRadtran with ALG is now possible with Ubuntu for Windows instead of the previous Cygwin.
  • The Atmosphere Generator Toolkit (v.1.0.1) has been updated to be compatible with NWP SAF atmospheric profiles database.
 01/09/2021

ARTMO (v3.29):

  • For SAIL, an option has been added to plot the loaded soil spectra. 
  • The Retrieval toolboxes have been updated (SI: v.1.27, MLRA: v.1.26; LUT inversion: 1.15).  Some bugs have been fixed and small utilities have been added. For instance, in the MLRA toolbox, PCA outputs and band ranking options have been added.
  • A new classification toolbox is introduced here (v1.00). It contains over 15 (machine learning) classifiers. Also a “LabelMeClass” tool has been added to collect labeled spectra (v1.00). 
  • For direct access to the latest complete latest version, access can be provided through a GitLab repository by request.

DATimeS (v1.12):

  • More reliable detection of heterogeneous spatial patterns of phenological indicators (i.e., crop key growth stages) throughout multiple seasons.
  • Solved an issue with Tiff images in specific cases.
  • New option to save the interpolation results in a .txt file for specific regions or areas.
  • Minor bugs have been corrected.

ALG (v3.1):

  •  ALG is now compatible with two new open-source models: ARTDECO and SBDART. Windows users can run these models using Ubuntu for Windows (see ALG Help/user guide).
  • ALG allows now to run your radiative transfer models on an external Linux server. Note: this functionality is not documented. Please contact us if you want to use it.
  • Minor bugs have been corrected, particularly occurring when porting ALG’s GUI to the newest Matlab R2020b app designer tool.
18/08/2022

ARTMO (v3.30):

  • SCOPE v.2.1 has been added.
  • The supervised machine learning classification algorithms toolbox (MLCA) has been improved (v.1.01) with per-class statistics, uncertainty, and band analysis.
  • The LabelMe (v.1.01) and LabelMeClass (v.1.01) tools have been improved with spectral data extraction options, e.g. based on a .txt file with GPS coordinates. 
  • A new Uncertainty module has been added to the MLRA toolbox (v.1.27); it allows to map uncertainties for any regression algorithm. New machine learning algorithms have been added. 
  • The Emulator toolbox has been expanded with an option to map uncertainty of an emulated scene (v.1.14).  Also new machine learning algorithms have been added.
  • Minor bugs have been corrected in LUT-based inversion (v.1.16), SI (v.1.28) and ARTMO essentials (v.3.30). 
19/09/2022

DATimeS has been updated as standalone version to 1.13

  • Phenological images are georeferenced.
  • Phenological parameters can be estimated using different thersholds in left and right side..
  • Now Phenological module works with big tiff files.
  • Debugged version.
 25/08/2023

ARTMO (v3.31):

  • The soil RT model MARMIT-2 has been added. Now RTM coupling is possible for soil-leaf-canopy-atmosphere RTMs.
  • The Retrieval and Classification toolboxes have been improved, e.g. with the possibility to process directly EnMAP or PRISMA images.
  • An Image Format Conversion tool has been added to convert EnMAP, PRISMA or Sentinel-2 formats to ENVI in the Retrieval toolboxes and LabelMe/LabelMeClass tools.
  • An option to create band settings of a regular sensor has been added to the Sensor too (v1.11).
  • Minor bugs have been corrected in LUT-based inversion (v.1.18), MLRA (v.1.28) Emulator toolbox (v.1.16) and ARTMO essentials (v.3.31). 

ALG (v3.2):

  • RAMI4ATM tool automates the generation and processing of atmospheric RTM data ready for the RAMI4ATM activity.
  • A Surface spectral library tool is implemented to allow users to include their own Lambertian and BRDF models for the TOA radiance execution mode.
  • The updated TOA radiance mode allows users to use built-in BRDF models within each atmospheric RTM.
  • Distribution of LUT points can now be done based on natural distributions, including correlations from existing atmospheric/climatological datasets.
  • LUTs can now be generated at a sensor spectral configuration.
  • The Atmosphere Generator tool now interfaces with the HITRAN database using HAPI (this is still ongoing so be aware of potential errors).