Feature Calculator (BETA)

Calculate numerical features of an image, the regions in a labeling (= segments) or segment on an image.

Please note: This node is in a beta-testing phase. We would be happy if you would report any bugs in our forum (https://tech.knime.org/forum/knime-image-processing).


Feature Set Configuration

Select Image Column
Select the image column on which the features should be computed.
Select Labeling Column
Select the labeling column on which the features should be computed.
Column Creation Mode
Mode how to handle the selected columns. The processed columns can be added to a new table or appended to the end of the input table.
Select the dimensions on which the selected feature sets should be computed. If the input has three dimensions but only two dimensions are selected the node will slice the input and compute the features on the two dimensional slices.
Select the feature set to compute and add it with the "Add" button to the "Selected Feature Sets" - Panel.
You can add as many feature sets as you want. If you add feature sets which are not compatible with the input columns or the selected dimensions the node will skip the feature set during execution.
Selected Feature Sets
Some of the feature sets can be configured. You can select the features which should be computed or enter other parameters for the feature computation.

Label Segment Settings

Append labels of overlapping segments
If true, the labels of overlapping segments will be appended in an additional column (';'-separated).
Overlapping segments do NOT need to completely overlap
If selected, segments will also be counted as overlapping, if so don't overlap completely.
Append segment information
If selected, the bitmask and the label of each segment will be added.
Segment Label Filter
Only labels which hold the given regular expression are considered during the calculation.

Input Ports

Input table with at least one image or one labeling column.

Output Ports

Feature table.


This node has no views


Further Links


You want to see the source code for this node? Click the following button and we’ll use our super-powers to find it for you.