Remote Sensing Field Work Tools

In remote sensing, designing a field campaign to collect ground-truth data can be a challenging task. We need to collect representative samples while accounting for issues such as budget constraints and limited accessibility created by e.g. poor infrastructure. As suggested by Olofsson et al. (2014) , this demands the establishment of best-practices to collect ground-truth data that avoid the waste of time and funds. 'fieldRS' addresses this issue by helping scientists and practitioners design field campaigns through the identification of priority sampling sites, the extraction of potential sampling plots and the conversion of plots into consistent training and validation samples that can be used in e.g. land cover classification.

CRAN version CRAN downloads CRAN downloads

fieldRS is an R package to support Remote Sensing oriented field work. It was developed at the Department of Remote Sensing of the University of Würzburg, Germany, within the Central Asia Waters (CAWa) project funded by the German Federal Foreign Office. It is part of the German Water Initiative for Central Asia (the so-called Berlin process), launched by the German Federal Foreign Office launched on the 1st of April of 2008 to support peaceful, sustainable and mutually beneficial management of transboundary water resources in Central Asia.

fieldRS was designed to ease the collection and management of ground truth data. It provides tools that help select sampling sites, correct and pre-process training data and preview classification results in order to identify potentially incorrect samples. The package provides a tutorial (found here) which contains a detailed methodology based on the package. It guides potential users in the full process of designing a remote sensing field campaign from the selection of potential study sites to the development predictive models. At the moment, fieldRS is oriented towards land cover mapping but will be

What else are we doing?

The Department of Remote Sensing of the University of Würzburg has developed other R packages that might interest you:

Click here for news on our department.

Supported by




fieldRS 0.2.1



  • extractFields() offers multiple drawing options and includes smoothing
  • ecDistance now accepts matrices and data.frames
  • labelCheck() corrects labels automatically given a reference vector
  • poly2sample() preserves the unique identifiers of each polygon
  • ccLabel() allows positive/negative thresholds (informs if the output should be higher or lower than the specified value)


  • "class_count" replaced by "richness" in "priority" argument within rankPlots()
  • "diversity" added to "priority" argument within rankPlots()


fieldRS 0.1.2



  • Added functions simpleTrace() and smoothTrace()


  • Update vignette
  • labelCheck() now returns correct sample frequencies


  • extractFields() provides multiple methods to draw polygons


fieldRS 0.1.1



  • Added functions poly2Sample() and checkOverlap()


  • Update vignette formatting


fieldRS 0.1.0


Initial release to CRAN with the following functions:

  • spCentroid()
  • relative.freq()
  • raster2sample()
  • poly2sample()
  • rankPlots()
  • mape()
  • labelCheck()
  • geCheck()
  • extractFields()
  • erosionFilter()
  • ecDistance()
  • derivePlots()
  • classModel()
  • ccLabel()

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.2.3 by Ruben Remelgado, 9 months ago

Report a bug at

Browse source code at

Authors: Ruben Remelgado [aut, cre]

Documentation:   PDF Manual  

GPL (>= 3) license

Imports raster, sp, caret, ggplot2, grDevices, geosphere, vegan, rgeos, stringdist, concaveman

Suggests knitr, rmarkdown, kableExtra, imager, randomForest, rgdal, RStoolbox, foreign, maptools

See at CRAN