This function can be used to aid in classifying spectrally diverse classes by splitting the input classes into subclasses using a clustering algorithm. After classification, these subclasses are merged back into their original parent classes. For example, the training data for an agriculture class might have both fallow and planted fields in the training data, or fields planted with different crops that are spectrally dissimilar. This function can be used to automatically split the agriculture class into a number of subclasses. The classifier is then run on this larger set of classes, and following classification, these subclasses can all be merged together into a single overall agriculture class.
1 | split_classes(train_data, split_levels, verbose = FALSE)
|
train_data |
a |
split_levels |
(optional) a list giving the names of the levels to split. If missing, all levels will be split. |
verbose |
whether to report status while running |
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.