You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi,
I'm testing mialsrtkTVSuperResolution with infant rather than fetal images. I've made modifications to support non slice based registration, as we don't have the same motion problems. This is low field data with multiple acquisitions with thick slices, but orthogonal orientations.
At the moment I'm struggling to get the LOO PSNR estimates. I've found mialsrtkEvaluateReconstructionQualityMeasures and can get it to run in the situation where the first image in the stack is removed from the TVSuperResolution step, but segfaults otherwise (i.e. when one of the other images is removed).
I think my problem is that the image left out has not been aligned to the final reference, for example using transforms created by mialsrtkImageReconstruction.
So, the question is whether mialsrtkEvaluateReconstructionQualityMeasures is the appropriate tool, and whether there is a utility for applying the slicewise transforms to allow me to test whether my theory correct.
Failing that, is there a script somewhere that computes the PSNR curves from the paper?
Thanks
The text was updated successfully, but these errors were encountered:
Hi,
I'm testing mialsrtkTVSuperResolution with infant rather than fetal images. I've made modifications to support non slice based registration, as we don't have the same motion problems. This is low field data with multiple acquisitions with thick slices, but orthogonal orientations.
At the moment I'm struggling to get the LOO PSNR estimates. I've found
mialsrtkEvaluateReconstructionQualityMeasures
and can get it to run in the situation where the first image in the stack is removed from the TVSuperResolution step, but segfaults otherwise (i.e. when one of the other images is removed).I think my problem is that the image left out has not been aligned to the final reference, for example using transforms created by
mialsrtkImageReconstruction
.So, the question is whether
mialsrtkEvaluateReconstructionQualityMeasures
is the appropriate tool, and whether there is a utility for applying the slicewise transforms to allow me to test whether my theory correct.Failing that, is there a script somewhere that computes the PSNR curves from the paper?
Thanks
The text was updated successfully, but these errors were encountered: