Hi,
In MLT the documentation (hover info) says that the threshold is based on "sigma units" ... However, sigma is not defined. It appears to me it is not standard deviation... and likely some other scheme.
Does anyone know the answer?
(MMT does appear to use the mean absolute deviation about the median... it does not appear to be the same "MAD" reported in statistics since the values in a noise reduced layer are slightly different than the *median* absolute deviation about median.)
-adam
