added spm scheme

parent e91a8eb1
......@@ -7,6 +7,13 @@
Package provides easier access to EU environmental maps and functions to produce and improve new value-added spatial layers. Key functionality includes:
* `train.spm` --- train a spatial prediction model using [mlr3 package]( implementation with spatial coordinates and spatial cross-validation,
* `accuracy.plot` --- plots predicted vs observed values based on the result of `train.spm`,
* `extract.tif` --- extracts points with space-time observations from a list of tifs with different begin / end reference periods,
* `predict.spm` --- can be used to predict values from a fitted spatial prediction model at new locations,
A general tutorial for `train.spm` is available [here]( The eumap package builds up on top of the [mlr3](, [terra]( and similar packages. As such, it's main purpose is to automate as much as possible Machine Learning and prediction in a scalable system.
![General workflow eumap package](../img/spm_general_workflow.png)
Warning: most of functions are optimized to run in parallel by default. This might result in high RAM and CPU usage.
......@@ -6,14 +6,14 @@
Tiling system with the cartographic and image coordinates. Can be used to speed up processing of the data.
The \code{eugrid30km} is an object of type \code{SpatialPolygonsDataFrame} with the following columns:
\item{\code{xl}}{numeric; lower left x-map coordinates},
\item{\code{yl}}{numeric; lower left y-map coordinates},
\item{\code{xu}}{numeric; upper right x-map coordinates},
\item{\code{yu}}{numeric; upper right y-map coordinates},
\item{\code{offst_y}}{numeric; offsets in the image coordinates lines (starting from top left)},
\item{\code{offst_x}}{numeric; offsets in the image coordinates rows (starting from top left)},
\item{\code{rgn_dm_y}}{numeric; tile dimensions in image coordinates},
\item{\code{rgn_dm_x}}{numeric; tile dimensions in image coordinates},
\item{\code{xl}}{numeric; lower left x-map coordinates}
\item{\code{yl}}{numeric; lower left y-map coordinates}
\item{\code{xu}}{numeric; upper right x-map coordinates}
\item{\code{yu}}{numeric; upper right y-map coordinates}
\item{\code{offst_y}}{numeric; offsets in the image coordinates lines (starting from top left)}
\item{\code{offst_x}}{numeric; offsets in the image coordinates rows (starting from top left)}
\item{\code{rgn_dm_y}}{numeric; tile dimensions in image coordinates}
\item{\code{rgn_dm_x}}{numeric; tile dimensions in image coordinates}
......@@ -29,6 +29,14 @@ This animation shows the land-cover classes for an area located in Sweden (tile
![pyeumap Workflow](img/land_cover_predictions.gif)
Spatiotemporal Machine-Learning
In the Geo-harmonizer project, we prepare Analysis-Ready Earth Observation images from Landsat and
Sentinel missions, then use ground observations from the European Commission projects such as [LUCAS surveys]( and CORINE and similar to
overlay the ground observations in the spacetime cubes. From this data we create spatiotemporal regression and classification matrices (see: [sample data set]( The eumap package (python and R versions) will allow accessing this data and testing models that apply Machine Learning for predictive mapping in spacetime.
![EU LUCAS scheme](img/Scheme_LUCAS_poinst_st_overlay.png)
© Contributors, 2020. Licensed under an [Apache-2]( license.
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment