SCI Seg3D Mailing List

Text archives Help

Re: [Seg3D] Automatic Segmentation

Chronological Thread 
  • From: Jess <>
  • To: "" <>
  • Cc: "" <>
  • Subject: Re: [Seg3D] Automatic Segmentation
  • Date: Mon, 15 May 2017 10:27:00 -0600

Hi Festim,

There are a few general things that can help you get started. First about the different pieces that you may need:

The provenance window will show you the steps to recreate a layer.  It allows you to focus on processing the data and get all the steps afterward.  You will probably need to optimized the steps for scripting.  

The controller window allows you to see al the function calls that seg3d makes.  The newest version will also show the python syntax that you can copy to a script file.  There is also a state variables tab which will show all the state variables that are exposed and that you can change in the script (using set and get functions).  You can also use the controller window to verify your script as you write it.  

The python console is where you enter python function calls or run the python script that you write.  You will need a basic understanding of python to use this.

For more information on these windows and functions, please see the relevant sections in the basic functionality guide (starts on page 25 of the pdf):
There are also some python code examples that you can use to get the hang of the process:
There was also a recent thread where we talk about a similar applications of python scripting:

And as Paul said,  we can give you more specific answers with more specific questions.  Please try these out and come back with any other questions you have.  


On May 15, 2017, at 8:26 AM, Paul Holcomb <> wrote:


Could you please provide more detail on the type of images you're using (MRI, CT, light microscopy, etc.) and what you're trying to segment?  That would help our community give you the best advice possible.


Paul Holcomb

Hello Dear colleagues

I hope this e-mail finds you all well.

I am a Biomedical master student studying in Germany and I am currently working on a project with Seg3D. My goal is to come up with a way to do automatic segmentation with possibly as little input from users as possible.

So far I understand that one needs to use Provenance, the Python and Controller windows for this task but I have no more info on this. 

As I am in the very starting point of this task, I would like to ask if any of you can give me some more information, hints or directions for me to achieve the goal.

Thank you in advance

Festim Xhoxhaj

Master student of Biomedical Engineering

Anhalt University of Applied Sciences​

Seg3d mailing list:
To unsubscribe, email with the "unsubscribe seg3d" in the message body.
Seg3d mailing list:
To unsubscribe, email with the "unsubscribe seg3d" in the message body.

Archive powered by MHonArc 2.6.18.

Top of page