1
$\begingroup$

In the research of the computed tomography (CT) reconstruction, one needs to simulate the noise during CT projection and capturing. Then, my questions are:

  1. How about the proper noise type?
  2. How to add the noise, in the sinogram domain or the frequency domain?

Any reference materials are also preferred.

| improve this question | |
$\endgroup$
2
$\begingroup$

The easiest, and probably most straight forward way, seems to add the noise in the measurement domain, hence the sinogram.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Thank you. how about the noise type? Gaussian? $\endgroup$ – lxg Sep 27 '16 at 16:47
  • 1
    $\begingroup$ I don't know. I wouldn't expect Gaussian noise, since your detector cannot measure negative signals. I would expect it to be a Poisson-distribution. $\endgroup$ – M529 Sep 27 '16 at 18:25
  • $\begingroup$ thank you, can you recommend any references? $\endgroup$ – lxg Sep 27 '16 at 23:14
  • $\begingroup$ No, I can't. I would take a first step by googling noise computer tomography ;) $\endgroup$ – M529 Sep 28 '16 at 17:08
  • $\begingroup$ Thanks. I found a similar question: cn.mathworks.com/matlabcentral/newsreader/view_thread/315689 $\endgroup$ – lxg Sep 29 '16 at 23:09

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.