[Rtk-users] Fwd: About the newspacing flag

Sat Feb 20 08:50:03 EST 2016

Hi Guangming,

In many image file format the pixel size (spacing) can be stored within the
file data. If the pixel size in the projection files are correct (e.g.
spacing=1 in your 256x256 P1, or spacing=2 in your 128x128 P2), then you
can use them without giving the newspacing flag. The newspacing flag just
gives you a chance to force RTK to use the pixel size you specified instead
of the one that stored in the image files.

I am not into rtkSART but I guess the forward projection works with the
input grid, so 128x128 if the input projections are 128x128. Please correct
me if I am wrong.

Regards,
Chao

2016-02-20 14:28 GMT+01:00 Guangming Zang <guangming.zang at kaust.edu.sa>:

> Hi Chao,
> Thanks very much for your kind help. i see, now i know there is no
> upsampling at all.
> But i am still a little confused about "then you do not even need to use
> the newspacing flag" . Since if i want to use same geometry(sid, sdd,
> fixed detector size) after dowmsampling (from P1 to P2).
> The size of each projection image changes from 256^2 to 128^2. so to
> maintain the same detector size, the spacing must be 2, i.e, the newspacing
> should set to 2. Is that correct???
> Actually, the question i have now is that , if before the downsample, each
> image size is 256^2. and in forward projecting part in SART alglrothm, say
> the number of ray generated is 256^2.
> Then after downsampling,i.e., when using P2 as input for rtkSART, what is
> the number of ray generated in forward projections part?? is it 128^2 since
> the number of pixel is 128^2 ???
> Thanks very much!
> Regards Guangming
>
> On Sat, Feb 20, 2016 at 4:17 PM, Chao Wu <wuchao04 at gmail.com> wrote:
>
>> Hi Guangming,
>>
>> There is no upsampling at all. It simply overwrites the pixel size of the
>> projection images stored in the image files. If for instance when you
>> downsampled your P1 to P2 with a factor 2, you also changed the pixel size
>> from 1 to 2 already, then you do not even need to use the newspacing flag.
>>
>> Regards,
>> Chao
>>
>>
>> 2016-02-19 23:50 GMT+01:00 Guangming Zang <guangming.zang at kaust.edu.sa>:
>>
>>> Hi Simon,
>>> Thanks for your kind reply and information provided.
>>> Basically the problem i encountered is that without changing any
>>> geometry (i.e., sid, sdd, detector physical size), if i downsample my
>>> original projection images, how can i reconstruct corresponding volume
>>> correctly.
>>> In RTK, it seems that i just need to change the input projection file
>>>  and newspacing parameter (as same as the factor of downsample), then i got
>>> promising results. but in my own application, i simulate the situation, but
>>> got some aliasing. So i really want to know how it work behind our RTK.
>>> I thought that this parameter is to simulate the detector's spacing size.
>>> And in my experiment, i set the detector physical size as 256mm*256mm,
>>> and i got projection images P1.mha with size of 256^2 , now i downsample
>>> these images with factor of 2( i.e., the images size change from 256^2 to
>>> 128^2 ), from which i got P2.mha
>>> in our rtkfdk, after the dowmsample. i just need change input
>>> projections from P1.mha to P2.mha, besides, i need change flag newspacing
>>> from 1 to 2. both of the results from RTK  are pretty good.
>>> my questions are:
>>> 1) can i just regard the newspacing as the upsample operation with
>>> nearest neighbor filter??
>>> 2) how many rays RTK generated in forward projecion (i.e. ray tracing
>>> part ) for each image in P2.mha??? is it 256^2????
>>>
>>> To sum up,in my own application, can i reproduce the result from RTK for
>>> flag "newspacing as 2"  by doing:
>>> upsample image by factor of 2 (got a 256mm*256mm image ) and just
>>> generate half number of rays in raytracing (forward priject ) part???
>>>
>>> Thanks for your time and help in advance,
>>> -Regards Guangming
>>>
>>>
>>>
>>> On Sat, Feb 20, 2016 at 1:15 AM, Simon Rit <
>>> simon.rit at creatis.insa-lyon.fr> wrote:
>>>
>>>> Hi Guangming,
>>>> The parameter is processed in rtkGgoFunctions
>>>> <https://github.com/SimonRit/RTK/blob/master/code/rtkGgoFunctions.h>.h.
>>>> It's passed to an itk::ChangeInformationFilter
>>>> <http://www.itk.org/Doxygen/html/classitk_1_1ChangeInformationImageFilter.html>
>>>> at the beginning of the rtk::ProjectionsReader.h
>>>> <http://www.openrtk.org/Doxygen/classrtk_1_1ProjectionsReader.html>.
>>>> But in the end, there is nothing to explain, it's simply changes the
>>>> spacing of the input projections.
>>>> I hope this helps,
>>>> Simon
>>>>
>>>> On Fri, Feb 19, 2016 at 8:52 PM, Guangming Zang <
>>>> guangming.zang at kaust.edu.sa> wrote:
>>>>
>>>>>
>>>>> ---------- Forwarded message ----------
>>>>> From: Guangming Zang <guangming.zang at kaust.edu.sa>
>>>>> Date: Fri, Feb 19, 2016 at 10:49 PM
>>>>> Subject: About the newspacing flag
>>>>> To: rtk-users-request at public.kitware.com, Chao Wu <wuchao04 at gmail.com>,
>>>>> Simon Rit <simon.rit at creatis.insa-lyon.fr>, "
>>>>> rtk-users at public.kitware.com" <rtk-users at public.kitware.com>,
>>>>> rtk-users at openrtk.org
>>>>>
>>>>>
>>>>> Dear RTK community,
>>>>> i have a question about the parameter "newspacing " in rtkFDK and
>>>>> rtkSART algorithm.
>>>>> it says:--newspacing  new spacing of input projections (before
>>>>> pre-processing).
>>>>>  I checked in the code, and it seems not  easy to find the correct
>>>>> part in the code about the newspacing .
>>>>> So could anyone  help me and explain about this parameter?? i want to
>>>>> simulate this preprocess step in my own project and compare to our RTK.
>>>>> -best regards Guangming
>>>>>
>>>>> --
>>>>> *Guangming Zang (Alex)*
>>>>> *King Abdullah University of Science and Technology(KAUST)*
>>>>> *University of Chinese Academy of Sciences(UCAS)*
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> --
>>>>> *Guangming Zang (Alex)*
>>>>> *King Abdullah University of Science and Technology(KAUST)*
>>>>> *University of Chinese Academy of Sciences(UCAS)*
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> ------------------------------
>>>>> This message and its contents, including attachments are intended
>>>>> solely for the original recipient. If you are not the intended recipient or
>>>>> have received this message in error, please notify me immediately and
>>>>> delete this message from your computer system. Any unauthorized use or
>>>>> distribution is prohibited. Please consider the environment before printing
>>>>> this email.
>>>>>
>>>>> _______________________________________________
>>>>> Rtk-users mailing list
>>>>> Rtk-users at public.kitware.com
>>>>> http://public.kitware.com/mailman/listinfo/rtk-users
>>>>>
>>>>>
>>>>
>>>
>>>
>>> --
>>> *Guangming Zang (Alex)*
>>> *King Abdullah University of Science and Technology(KAUST)*
>>> *University of Chinese Academy of Sciences(UCAS)*
>>>
>>>
>>>
>>>
>>> ------------------------------
>>> This message and its contents, including attachments are intended solely
>>> for the original recipient. If you are not the intended recipient or have
>>> received this message in error, please notify me immediately and delete
>>> this message from your computer system. Any unauthorized use or
>>> distribution is prohibited. Please consider the environment before printing
>>> this email.
>>>
>>> _______________________________________________
>>> Rtk-users mailing list
>>> Rtk-users at public.kitware.com
>>> http://public.kitware.com/mailman/listinfo/rtk-users
>>>
>>>
>>
>
>
> --
> *Guangming Zang (Alex)*
> *King Abdullah University of Science and Technology(KAUST)*
> *University of Chinese Academy of Sciences(UCAS)*
>
>
>
>
> ------------------------------
> This message and its contents, including attachments are intended solely
> for the original recipient. If you are not the intended recipient or have
> received this message in error, please notify me immediately and delete
> this message from your computer system. Any unauthorized use or
> distribution is prohibited. Please consider the environment before printing
> this email.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://public.kitware.com/pipermail/rtk-users/attachments/20160220/e1f7bd32/attachment-0021.html>