Inconsistent projection centre for the Radon transform
Hi everyone, I recently found out that scikit-image has a bug (and has had for a long time, it seems) in the Radon transform module (radon/iradon, not related to the discreete versions). The problem manifests itself as images being shifted by going through a forward and inverse Radon transform. In fact, the following code would shift an image N pixels both horizontally and vertically: # image = np.array(...) for i in range(N): sinogram = radon(image) image = iradon(sinogram) These issues can also easily lead to poorer reconstruction quality than what the data in the sinogram allows. The purpose of this message is two-fold: 1. To ask for review on my PR addressing this bug: https://github.com/scikit-image/scikit-image/pull/596 . I realize that this PR is probably not the easiest to review; I guess this reflects that it was tricky for me to get right. However, to me this just makes a review seem all the more important. 2. Alert any users that have based results on this code to check them carefully for systematic errors. If the radon_transform module has been used for tomography simulations, it is quite likely that the results are affected. In the case of using the module for reconstructing tomography data obtained otherwise, the results may be affected and they may not; this will depend on the specifics of the situation. Jostein
Dear Jostein On Sat, Jun 22, 2013 at 10:31 AM, <jostein.floystad@gmail.com> wrote:
2. Alert any users that have based results on this code to check them carefully for systematic errors. If the radon_transform module has been used for tomography simulations, it is quite likely that the results are affected. In the case of using the module for reconstructing tomography data obtained otherwise, the results may be affected and they may not; this will depend on the specifics of the situation.
Thank you for bringing this issue to everyone's attention. Rigorous code review, which leads to the uncovering of such mistakes, is what makes open source science so compelling to me. I'll review your PR, so that we can get this fixed ASAP. Regards Stéfan
participants (2)
-
jostein.floystad@gmail.com
-
Stéfan van der Walt