Inverse Pole Figure (IPF) coloring paraview

1,587 views
Skip to first unread message

kasra momeni

unread,
Apr 13, 2016, 5:58:49 PM4/13/16
to dream3d-users
Hi,

I want to visualize the data generated in dream3d using preview with IPF coloring. I was wondering how should I do it in preview?

best

Mingyan Wang

unread,
Apr 15, 2016, 4:41:36 AM4/15/16
to dream3d-users
Hi Kasra,

After you run the Generate IPF Color filter, you should have an attribute called IPFColors.
To view this in ParaView, you can uncheck the 'Map scalars' under 'scalar coloring' in the 'Properties' window. Of course, you have to select the IPFColors to view first.

Hope this helps,
Mingyan

kasra momeni

unread,
Apr 15, 2016, 5:30:47 PM4/15/16
to dream3d-users
Hi,

I have tried it and it works for the Ni super alloy example. However, I have another set of data that is 2D and although components of the Euler angle are varying within the sample, when I plot the IPF coloring generated by Dream3D it shows a uniform color. I have attached the dream3D pipeline and the data files, and would be very thankful if you could make any comment.

Best
K
2D_Alison_v3.json
304L ss H6_4k Site 1 Map Data 1.zip

kasra momeni

unread,
Apr 15, 2016, 5:37:50 PM4/15/16
to dream3d-users
Here is also the map I got from the EBSD software.

Sean Donegan

unread,
Apr 16, 2016, 12:12:04 PM4/16/16
to dream3d-users
Hi Kasra,

I took a closer look at your data, and have noticed a few things that probably explain the issues you've been having:
  1. Don't convert the Euler angles from degrees to radians. Previously I think we had told you to try that, and normally that is the case for HKL data, which stores its angles as degrees.  However, it looks like the read H5EBSD filter is doing that conversion under the hood.  Try removing the convert angle representation filter from your pipeline.
  2. The image you posted is for Euler coloring, where it looks like each Euler angle has been mapped to an RGB channel.  This is fundamentally different from IPF colors.  If you want to reproduce this exact Euler coloring from DREAM.3D, it is possible in a couple steps.  First, read your data in using the read H5EBSD filter, but do not check the Use Recommended Transformations filter.  This keeps the orientations in the same reference frame as from the HKL software.  Then simply write out your data as a .dream3d file and open the corresponding xdmf file in ParaView.  Now, if you view the Euler angles, you should get the same map as in your image (note that the sample reference frames are different; if you want to get the exact same image, you have to rotate the image 180 degrees about the x axis).  I've attached an image that shows this coloring as written from DREAM.3D
It is important to be aware that coloring by the Euler angles is quite ambiguous.  Generally, IPF colors are preferred, since they much more clearly communicate information about the true orientations.  This is because Euler coloring may not give you a unique map between a particular color and a specific orientation.

hope that helps,

-Sean 
EulerColors.png

kasra momeni

unread,
Apr 21, 2016, 10:45:10 PM4/21/16
to dream3d-users
Thank you very much! I could reproduce the figure. However, I need to write the output as the INL file format which the corresponding filter requires additional information. I have prepared the pipeline which is attached. The misorientation tolerance was chosen to be 0 degrees to avoid grouping the data. However, the Euler Angles in the output obtained after applying segment features is different and its magnitude became very limited. 
I was wondering why the Euler angles change after applying these filters which should supposedly avoid altering these angles, and how we can prevent it.
best
2D_Alison_v3_short.json

Sean Donegan

unread,
Apr 29, 2016, 8:37:37 AM4/29/16
to dream3d-users
Hi Kasra,

Please refer to this post for information on the issues regarding Euler angles:


The procedure described in the above post should help the issues regarding the ranges of Euler angles.

-Sean

On Wednesday, April 13, 2016 at 5:58:49 PM UTC-4, kasra momeni wrote:

Umar Manan

unread,
Dec 19, 2019, 7:10:52 AM12/19/19
to dream3d-users
hello Kasra,

i am having the same issue which you had in 2016. i am just curious did you find the solution if yes, what was it?

Michael Jackson

unread,
Dec 19, 2019, 3:52:39 PM12/19/19
to Umar Manan, dream3d-users
Did you try any of the example pipelines? there is a specific example
that exports IPF Colors into a standard image file called EDAX IPF
Colors.

Are you trying to view the data in Preview or ParaView?
_________________________________________________________
Mike Jackson mike.j...@bluequartz.net
BlueQuartz Software www.bluequartz.net
Principal Software Engineer Dayton, Ohio
> --
> You received this message because you are subscribed to the Google Groups "dream3d-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to dream3d-user...@googlegroups.com.
> To view this discussion on the web visit https://groups.google.com/d/msgid/dream3d-users/e21a9f0d-37e3-4866-946f-01cbebcef849%40googlegroups.com.

Michael Jackson

unread,
Dec 20, 2019, 9:48:35 AM12/20/19
to Muhammad Umar, dream3d-users

You are looking for the IPF Legends. We have them premade inside the “Data” directory that comes with DREAM.3D. Look in Data/OrientationAnalysis/IPF_Legend/*.png. They are all like 2K x 2K pixels (so very large). You will need to know what Laue system that your data uses in order to select the proper Legend.

 

--

Michael Jackson | Owner, President

      BlueQuartz Software

[e] mike.j...@bluequartz.net

[w] www.bluequartz.net

 

From: Muhammad Umar <umar....@gmail.com>
Date: Friday, December 20, 2019 at 1:19 AM
To: Michael Jackson <mike.j...@bluequartz.net>
Subject: Re: Inverse Pole Figure (IPF) coloring paraview

 

Dear Michael, 

 

Thank you very much for timely response and the suggestion. i have used that example and it works fine, but the output image file does not have any key of ipf map for reference and better understanding the grains orientation.

 

secondly i have created my own virtual micro-structure (ferrite and cementite phases) for simulations by DAMASK. now i want to extract the IPF maps of individual phases alongwith a reference color key. My RVE has the IPF colors and they are visible in the paraview software but again the reference color key is missing. this is where i need your guidance now.

 

Thank in advance.

--

 

Engr. Muhammad Umar

Lab EngineerCell: +92 334 6130884www.kfueit.edu.pk




 

Reply all
Reply to author
Forward
0 new messages