You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I think image processing should be seen as a subset of signal processing (wikipedia agrees). That separation is probably due to signal processing being historically 1D, while we now have nD Euclidean SP and non-Euclidean SP on manifolds/graphs/homogeneous spaces/groups/etc. thanks to abstract harmonic analysis and the underlying group and representation theory. I'll make a PR if people agree.
The text was updated successfully, but these errors were encountered:
I'd prefer to keep these separate. xref gh-313 for full update on categories.
That separation is probably due to signal processing being historically 1D, while we now have nD Euclidean SP and non-Euclidean SP on manifolds/graphs/homogeneous spaces/groups/etc. thanks to abstract harmonic analysis and the underlying group and representation theory
This is fair from a mathematical point of view, but from a Python users and general scientist background point of view, those are things no one really wants to know about. As you say, the fields are historically separate - and they still are mostly separate for the average user. Plus I want to limit the maintenance work here (e.g., we have new icons for both of these in gh-369 now).
I think image processing should be seen as a subset of signal processing (wikipedia agrees). That separation is probably due to signal processing being historically 1D, while we now have nD Euclidean SP and non-Euclidean SP on manifolds/graphs/homogeneous spaces/groups/etc. thanks to abstract harmonic analysis and the underlying group and representation theory. I'll make a PR if people agree.
The text was updated successfully, but these errors were encountered: