That's not true! Haven't you seen any of movies where computer freaks make
fashion-magazine quality pics out of 100x100 pixels image from a parking
place video? <BG>
Sure, it's true: if we have $100 note, we cannot make more just by copying
it (ok, some guys do, but in most cases, not for a long time), the same
with images: we have certain ammount of information in the image, imagine,
this is a photo of a newspaper page, quality is good, and we can read the
text. Downsampling (according to methods used) throws away several pixels
and adjusts those which are located nearby to look similar to those deleted
- we got some image, where we can read titles of our newspaper page, but
there are blurred lines instead of regular text - that information is lost.
Upsampling will not return informaton which is thrown away (or even was not
there at all, if image quality is poor), that's it. Sampling algorithms can
make fine gradients instead of enlarged square-shaped spots of particular
colors, but they cannot add any _information_ to image. This is true for
bitmap images, vector graphics are different story.
Atvars
_________________________________
"William Robb" <[EMAIL PROTECTED]> wrote:
> > If this was so great why aren't lousy webcam shots "cleaned
> up" till one can see birthmarks
> > the model's mother never knew existed?? Eh?
>
> Downsizing images works pretty good, upsizing (adding pixels)
> doesn't.
> William Robb
>
-
This message is from the Pentax-Discuss Mail List. To unsubscribe,
go to http://www.pdml.net and follow the directions. Don't forget to
visit the Pentax Users' Gallery at http://pug.komkon.org .
