I am currently working on a project related to motion artifacts in MR images. I have some MR images and i'm extracting data from those images using different trajectories (e.g. Spiral, Radial etc) and reconstructing them using gridding algorithm. Now I need to analyze the quality of the reconstruction by comparing the reconstructed and the original image. What might be some good ways to comapre images under such scenario?
Asked
Active
Viewed 22 times
0
-
Mean square error is the standard comparison algorithm for this kind of stuff. You'll see PSNR suggested as well, but that is just a meaningless mapping of the MSE. – Cris Luengo Mar 17 '20 at 19:47
-
Thank you for your answer. I already have MSE on my mind. Just want to explore some other options before deciding which could be best for this problem. – shaurov2253 Mar 17 '20 at 20:13