[Rtk-users] Reconstruct Nikon XTH320 data

[Simon Rit]

I hadn't noticed the xtekct file but now that you mention it, your
pixel spacing is also completely wrong in the projection image.  With
this header file

ObjectType = Image
NDims = 3
BinaryData = True
BinaryDataByteOrderMSB = False
CompressedData = False
TransformMatrix = 1 0 0 0 1 0 0 0 1
Offset = -50 -50 0
CenterOfRotation = 0 0 0
AnatomicalOrientation = RAI
ElementSpacing = 0.2 0.2 1
DimSize = 500 500 3142
ElementType = MET_FLOAT
ElementDataFile = CT_121029a_quarter.raw

I observe a large improvement. You just have to be sure that you
understand well the rest of the geometry to obtain a good result I
believe.
Simon

On Fri, Sep 20, 2013 at 11:02 PM, M Miller <croow at yahoo.com> wrote:
>
> The sid/sdd values can be seen in the *.xtekct file at line 17:
> SrcToObject=68.4973907470703
> SrcToDetector=870.86
>
> Changing the .mhd file is a good suggestion. I didn't know the offset value was used by rtk. I had previously just guessed the proj_iso_* values until the output looked better.
> I never use the metaIO format myself, but I thought it would be easier to interface it with rtk than the raw count tiff files. I have two additional CT systems that output similar tiff files and I was hoping that if I can get the Nikon data to reconstruct using metaIO I could later address using tiffs as the input without the mhd conversion.
>
>
> Thanks
>
> --------------------------------------------
> On Fri, 9/20/13, Simon Rit <simon.rit at creatis.insa-lyon.fr> wrote:
>
>  Subject: Re: [Rtk-users] Reconstruct Nikon XTH320 data
>  To: "M Miller" <croow at yahoo.com>
>  Cc: "Marc Vila Oliva" <marc.vila-oliva at creatis.insa-lyon.fr>, "rtk-users at openrtk.org" <rtk-users at openrtk.org>
>  Date: Friday, September 20, 2013, 3:07 AM
>
>  Hi,
>  I have been looking at your data. You're right, it's
>  complete, Marc's
>  download probably stopped.
>  Clearly you have a geometry problem and it's going to be
>  hard for us
>  to solve it for you. Where did you get your sid / sdd from ?
>  I have
>  also noticed something weird, you use positive offsets in
>  your
>  projection images that you compensate for with negative
>  offsets in the
>  geometry file. I would advise you to center your projection
>  images
>  with negative offset ("Offset = -250 -250 0" in the mhd
>  file) and to
>  remove your projection offsets if the detector is centered.
>  Simon
>
>  On Thu, Sep 19, 2013 at 9:06 PM, M Miller <croow at yahoo.com>
>  wrote:
>  >> Do you really have 3142 projections? Have you
>  cropped the raw data?
>  >
>  > The data does have 3142 projections. The
>  CT_121029a_quarter.raw that I uploaded is 3142000000 bytes.
>  I downloaded it onto another computer and the size was also
>  ~3GB. I also viewed the downloaded data with Avizo and it
>  wasn't corrupted. Maybe your download was interrupted?
>  >
>  > I've uploaded another mhd/raw pair that is much smaller
>  (500x500x 500 projections) and may be easier to handle. This
>  projection set is equivalent to the 3GB set (*quarter.raw)
>  with the number of projections resized to 500.
>  >
>  > Thank you for your time.
>  >
>  >
>  >
>  > _______________________________________________
>  > Rtk-users mailing list
>  > Rtk-users at openrtk.org
>  > http://public.kitware.com/cgi-bin/mailman/listinfo/rtk-users
>