Trainable Weka Segmentation

Use the Trainable Weka Segmentation plugin included with Fiji to classify images. Requires a classification model file, which must be created directly with the plugin. Returns image segmentation or classification probability maps.

This command is only a wrapper for the Trainable Weka Segmentation plugin. If the plugin is not installed, the command will not work.

Input Parameters


Image files to segment.

Type: Image File (ij.ImagePlus)


Classification model file. This file can be created from the Trainable Weka Segmentation plugin by running the classifier training process, then selecting Save Classifier. See the plugin documentation for more information.

Type: Weka Model File

Save Probabilities

If true, save a map of classification probabilities. If false (default), save a segmentation mask.

Type: Logical (boolean)

RGB Model?

If true, treat the image as an 8-bit RGB stack for purposes of the Weka classifier. If false (default), keep the default image settings.

Type: Logical (boolean)



Image containing either a classification probability map or a segmentation mask, depending on the input value of Save Probabilities.

Type: Image File (ij.ImagePlus)