dagsvm_classification_predict: Classifies Images with a DAG-SVM.

In henningte/TraceIdentification: Identification of Fluorescent Traces in High-Resolution Aerial Images

Description

dagsvm_classification_predict takes a tuned DAG-SVM and the respective paramters as returned by tunedagsvm_extractoptimisedparametersandmodels and a directory name with images to classify and returns the images that were classified with the DAG-SVM. The construction of DAG-SVM is derived according to \insertCitePlatt.2000TraceIdentification

Usage

dagsvm_classification_predict(optimparam, optimmodels, directorydata)

Arguments

optimparam	A data.frame object corresponding to the first element of a list returned by tunedagsvm_extractoptimisedparametersandmodels with the tuned parameters of the DAG-SVM.
optimmodels	A list corresponding to the second element of a list returned by tunedagsvm_extractoptimisedparametersandmodels with the tuned models of the DAG-SVM.
directorydata	A character string representing the directory where the image files to be classified are stored. No other files must be in this directory.

Value

The function returns a list with an element for each image file in directorydata containing the classified image.

Source

The function relies on the R packages raster (\insertCiteHijmans.2017TraceIdentification) and e1071 (\insertCiteMeyer.2015TraceIdentification).

References

\insertAllCited

See Also

tunedagsvm, tunedagsvm_extractoptimisedparametersandmodels.

Examples

#

henningte/TraceIdentification documentation built on May 26, 2019, 6:50 p.m.