The experiments described in the previous section were repeated for shuffle neighbourhoods with radius 1 (Euclidean distance), 1.5, 2.1, and 3.7, to test the hypothesis that this would extend the range over which different degrees of mis-registration could be discriminated. It can observed, however, that there is no obvious gain when going past a certain neighbourhood size. This is proven numerically in Subsection IV-E.

Roy Schestowitz 2007-03-11