Hi
In Computer Vision, multi-scale representation of a 2-D signal is done by
embedding the signal into a one-parameter family of derived signals, the
scale-space. This is usually done by convolution with Gaussian kernels of
varying variances.
Does anyone know how this is done in Statistics? Are there any statistical
references on this topic?
Thanking you.
Dr Coomaren P Vencatasawmy
Swedish University of Agricultural Sciences
Remote Sensing Section
Umeå
Sweden
[log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|