0 ×

Trackmate Tracker

KNIME Image Processing - Tracking Framework version by University of Konstanz

The Linear Assignment Problem (LAP) trackers implemented here follow a stripped down version of the renowned method contributed by Jaqaman and colleagues[1]. We repeat here the ideas found in the reference paper, then stresses the differences with the nominal implementation. Particle-linking happens in two step: track segments creation from frame-to-frame particle linking, then track segments linking to achieve gap closing. The mathematical formulation used for both steps is linear assignment problem (LAP): a cost matrix is assembled contained all possible assignment costs. Actual assignments are retrieved by solving this matrix for minimal total cost. We describe first how cost matrices are arranged, then how individual costs are calculated (from http://fiji.sc/TrackMate).


Maximum Object Distance
Maximum distance between to objects
Tracking Dimension
Select the dimension which will be used for tracking (in most cases its Time).
Allow Splitting
If checked, the algorithm tries to detect splitting objects.
Max Distance (Splitting)
Maximum distance two splitting objects may have.
Allow Merging
If checked, the algorithm tries to detect merging objects
Max Distance (Merging)
Maximum distance two merging objects may have.
Allow Gap Closing
If checked, a missing object at a certain frame can be recovered using the following tracking frames.
Max Distance (Gap Distance)
Maximum distance two objects may have, such that their track can be recovered using gap closing.
Max Gap Size (Frames)
Number of frames which can be skipped for gap-closing.


Alternative Linking Cost Factor
Factor used to compute alternative linking costs (see http://fiji.sc/TrackMate for details).
Cutoff Percentile
Cut-Off Percentile (see http://fiji.sc/TrackMate for details).
Calculate Track features
Calculates various features for each track and makes them avaiable at the second output port.

Notable Features:
  • The Longest Gap describes the lenght (in time units) of the longest gap int the track.
  • The Complex points counted in the "Number of Complex Points" features are defined as follows. A Complex Point has several imidiate predecessors (like a merge) as well as several successors (like a split) regarding the surrounding time slices.
  • The Track Duration describes the total ammount of time the Track is visible.
  • The Track Start describes the first timepoint a track appears at.
  • The Track End describes the last timepoint a track appears at.
  • The Track Displacement describes the total distance traveled by the tracked object.
Calibration of the Features: (Customizable calibration will be avaiable in a future version).
  • Location based features: 1 = 1 mm
  • Time based features 1 = 1 sec
Attach Original Labelings
Attaches the original labelings to the tracks, so that each labeled pixel keeps his original labels.
Use a Custom Track Prefix
If a custom prefix for the tracknames is used instead of the default one "Track: ".
Custom Track Prefix
The prefix for the name of the tracks. If used together with Attach Original Labelings , make sure to enter a name that is not part of the name of any labeling.

Column Settings

Feature Columns Selection
Select the columns containing features of the spots, theses are used used to improve the tracking.
Other Column Selection
Select the columns containing the Bitmask, Labels, and the Source Labeling.

Input Ports

Table with cells to be tracked

Output Ports

Labeling with tracked cells
Features calculated over the Tracks


Table Cell View
Resulting Tracking

Best Friends (Incoming)

Best Friends (Outgoing)



To use this node in KNIME, install KNIME Image Processing from the following update site:

Wait a sec! You want to explore and install nodes even faster? We highly recommend our NodePit for KNIME extension for your KNIME Analytics Platform.


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.