SAMARA
is an iterative algorithm for 3D
reconstruction, which is more tolerant against input
data imperfections than the standard Feldkamp
method. Such data imperfections can have any
nature, e.g.
-
absorption non-linearities
-
scanner geometry misalignment
-
non-equidistant angular stepping
-
insufficient number of projections (e.g. less than 200).
The
technique is designed to reconstruct better
images than a standard Feldkamp
procedure. |