The R-package scapesClassification
is designed to translate users’
mental models of seascapes, landscapes and, more generally, of
geo-spaces into computer representations (classifications). Spaces and
geographic objects are classified with user-defined rules taking into
account spatial data as well as the spatial relationships existing among
different classes and objects.
Landscapes and seascapes tend to have prominent features easy to identify. These features can be considered as anchor locations, locations around which a classification process can start and evolve.
A classification process can take into account the spatial relationships that are expected to exist among different classes, i.e., where a segment of space or an object is expected to be in relation to other segments and objects.
Based on such relationships, it is possible to estimate where a certain class is expected to exist and to perform focal evaluations of classification rules: rules are only evaluated at suitable locations, thus, limiting possible misclassification cases.
A classification process is seen as multi-step: as new portions of a raster are classified they can be used to define new focal areas over which classification rules are evaluated.
If you are just getting started with scapesClassification and you would like to have a general overview of the package capabilities you can consult the github page and the working example articles. For a deeper understanding of how the package works you can consult the implementation articles and the examples throughout the package documentation.
You can install the released version of scapesClassification
from
CRAN with:
install.packages("scapesClassification", dependencies = TRUE)
And the development version from GitHub with:
# install.packages("devtools")
devtools::install_github("ghTaranto/scapesClassification", dependencies = TRUE)
Any scripts or data that you put into this service are public.
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.