|
Re: Resampling sharp Foveon images
In reply to Roland Karlsson,
6 months ago
|
There are several issues. It is easy to show that an image cannot be both "spatial limited" and "band limited." It's easy to prove since the Fourier transform of an "spatial limited" function (image) is an analytic function hence it can not be zero in an open set unless it is zero everywhere which means the original image is zero everywhere.
While the reconstruction is not "perfect" it can be made quite close to perfect. Take for example a CD. The music that has been digitized on the DC is turned back to analog for listening and not song last forever. The reconstruction is quite faithful - the error between the original and the reconstruction is on the order of -80 dB.
Roland Karlsson wrote:
I guess pure sinc reconstruction in practice is useless for images.
So - in practice - the prerequisites for the sampling theorem is not met.
--
/Roland
X3F tools:
http://www.proxel.se/x3f.html
https://github.com/rolkar/x3f
--
Truman
www.pbase.com/tprevatt
| Post (hide subjects) | Posted by | When | |
|---|---|---|---|
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago | |||
| 6 months ago |