Getting started with Theba

This guide tells you how to take Theba for a test drive.


Check if you have Java 1.5 or later installed, and if you don't get the latest Java-version from Sun (here). You need a machine with 512MB ram or more (to give Java 384MB of them) to run Theba. If you have less than the recommended amount of ram, the watershed operation and other features may not work properly. Also note that if you use a bigger dataset than the included sample, the memory requirements will increase. Theba requires 200MB free disk space on the drive you install it, more if you use bigger datasets than the included sample (it caches the dataset).


Download the latest release (1.0.4 at the time). Unzip it. You will get a directory named theba-1.0.4 (or similar). Enter that directory. It contains a Jar-file, documentation and a sample data set.

Execute it

The Jar file is what to execute.

If you got 512MB or more memory (recommended): just double-click the theba jar-file.
Otherwise, execute it using 'java -Xmx128 -jar theba.jar' from the command line. If you have less than the recommended amount of ram, the watershed operation and other features may not work properly.

Execute it using 'java -Xmx512m -jar theba.jar' from the command line. Make sure 'java' is Sun Java. Here too, change the -Xmx setting if you have less than 384MB RAM available to Java, to for example -Xmx128m.

Load a dataset

Theba reads raw images. By 'raw' we mean that they are datafiles without any metadata. That means you have to know the dimensions of the data (e.g. 100 images of resolution 1024x256 where each data point uses one byte) in order to read it. A sample dataset is included. Look for it in the sampledata/ directory that is included with theba-releases.

Select a tracker

Theba supports multiple trackers through a plugin-architecture. Select the recommended 'FiberTracker' if you are uncertain what to choose. For a brief overview of the trackers, see the front page.

Press 'OK' to load the selected dataset and tracker!

Theba ready!

Once the dataset has finished loading, you are greeted by the screen below:

For instructions on usage, take a look at the documentation. A quick summary follows below:

The interface should be pretty intuitive, you select lumens (holes in fibers) as seeds and then track the lumens and from there you can track the fiber walls. Also, there is a number of measurements you can do on regions (3d) and areas (2d) in addition to plugins for manipulating the whole volume, like watershed and dilate.

To give it a spin try the following:

Use Theba for development

You are free to use the Theba source code like you wish. Check out the javadoc documentation (html (browseable), html (zipped), pdf)