[Rtk-users] detector with not multiple pixel gap

[Simon Rit]

Hi Julia,
The artifact on the first image is not obvious to me. Can you clarify which
artifact is visible on the first image but not on the second?
In the second image, is the gap exactly at the center of the image? If yes,
I don't see a better solution than interpolating between adjacent pixels.
Simon

On Mon, Dec 7, 2020 at 6:11 AM Julia Semyakishkina <
prekrasnaya1985 at gmail.com> wrote:

> Hi!
> I have a slightly non-typical task. I've got a detector with a vertical
> gap that is not equal to pixel, so if I just insert interpolated pixel(s)
> to projection then corresponding geometry artifacts appear.
> https://i.imgur.com/1gwW1vx.png
>
> In order to take full account of the gap I'm trying to reconstruct my
> volume twice. So there are two sets of projections on opposite sides of the
> gap. I passed simple steps:
> - firstly reconstruct volume for the left side with correct offset that
> takes into account the gap;
> - reconstruct volume for the right side with also correct offset and
> initial volume from previous reconstruction;
>
> The volume obtained by this sequence does not suffer from geometry
> artifacts! However, there is a bright point in the center of the volume and
> the lines that come out of it. https://imgur.com/nUJ1IW5. Obviously, it's
> related to projection's edge, just like a fov circle in usual
> reconstruction. Is there any way to get rid of it ?
> _______________________________________________
> Rtk-users mailing list
> Rtk-users at public.kitware.com
> https://public.kitware.com/mailman/listinfo/rtk-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://public.kitware.com/pipermail/rtk-users/attachments/20201208/61cb389e/attachment.htm>