Subpixel accuracy analysis of phase correlation registration methods applied to aliased imagery

Phase Correlation is an efficient technique for subpixel image alignment under translational motion, but its subpixel accuracy normally suffer because of aliasing, which can be found in some undersampled imaging systems. In this work we review, compare, and mix different phase correlation based methods and test them under diverse aliasing conditions, from partial to total. Results show that with some additional features, the methods based on the one or two dimensional Gaussian fit of the spatial phase correlation peak are the less sensitive to severe aliased imagery. copyright by EURASIP.