«

Suppose you've taken a photo that is blurred (not due to
movement, but due to improper focus).
Would it be possible, if you could look through just the right
type of lens, to see the picture correctly focused?
If so, would it be possible for software to calculate the
focus problem, or even for a photographer to just go through
progressively different foci to fix something that is blurred?

On Jul 12, 4:04*pm, Scotius <yodas...@mnsi.net> wrote:
> * * * * Suppose you've taken a photo that is blurred (not due to
> movement, but due to improper focus).
> * * * * Would it be possible, if you could look through just the right
> type of lens, to see the picture correctly focused?
> * * * * If so, would it be possible for software to calculate the
> focus problem, or even for a photographer to just go through
> progressively different foci to fix something that is blurred?

Google 'deconvolution' and/or 'richardson-lucy'

To answer your question simplistically.. No. *Some* processing
techniques can recover *some* information, but in essence, once the
detail is blurred into other detail, you can never be quite sure that
what you recover is real.. Typically, deconvolution introduces
artefacts along with any 'recovered' detail.

It's a bit like enlarging - there's no free lunch - if you want real
detail, you must capture it adequately in the first place.

"Scotius" <> wrote in message
news:...
> Suppose you've taken a photo that is blurred (not due to
> movement, but due to improper focus).
> Would it be possible, if you could look through just the right
> type of lens, to see the picture correctly focused?
> If so, would it be possible for software to calculate the
> focus problem, or even for a photographer to just go through
> progressively different foci to fix something that is blurred?

I very muvh doubt it, and I thought adaptive optics was used
to partialy overcome atmospheric interference. This is employed in ground
based networked telescopes both for visible light and radi wave and IR I
think.It's used by adapting the shape of the 'reflector' or whatever is
grabbing the data from teh objects

On 15/07/2010 13:51, whisky-dave wrote:
> "Scotius"<> wrote in message
> news:...
>> Suppose you've taken a photo that is blurred (not due to
>> movement, but due to improper focus).
>> Would it be possible, if you could look through just the right
>> type of lens, to see the picture correctly focused?

In general "the right type of lens" cannot be physically realised.

>> If so, would it be possible for software to calculate the
>> focus problem, or even for a photographer to just go through
>> progressively different foci to fix something that is blurred?

Software solutions exist for deconvolution of images with a uniform or
nearly uniform well characterised point spread function. Methods used to
make the Hubble images before its myopia was corrected for instance.

And there are clever coded aperture imaging arrays where the data
collected is used to infer an image with better light grasp and much
larger depth of field than a conventional circular lens aperture. eg

http://www.paulcarlisle.net/old/codedaperture.html
>
> I very muvh doubt it, and I thought adaptive optics was used
> to partialy overcome atmospheric interference. This is employed in ground
> based networked telescopes both for visible light and radi wave and IR I
> think.It's used by adapting the shape of the 'reflector' or whatever is
> grabbing the data from teh objects

There is a "rubber" mirror somewhere in the imaging train that is used
to adjust the wavefronts to obtain the sharpest possible rendition of an
artificial or real guide star. It is able to take out some of the
atmospheric seeing on big scopes.

Cheap versions to take out tip-tilt errors are available to amateur
astronomers and for small fields of view like planets the humble webcam
coupled with software allows keen amateurs to get images that would be
better than top observatories could manage a few decades ago.

Regards,
Martin Brown

On Thu, 15 Jul 2010 13:51:52 +0100, "whisky-dave"
<whisky-> wrote:

>
>"Scotius" <> wrote in message
>news:.. .
>> Suppose you've taken a photo that is blurred (not due to
>> movement, but due to improper focus).
>> Would it be possible, if you could look through just the right
>> type of lens, to see the picture correctly focused?
>> If so, would it be possible for software to calculate the
>> focus problem, or even for a photographer to just go through
>> progressively different foci to fix something that is blurred?
>
>I very muvh doubt it, and I thought adaptive optics was used
>to partialy overcome atmospheric interference. This is employed in ground
>based networked telescopes both for visible light and radi wave and IR I
>think.It's used by adapting the shape of the 'reflector' or whatever is
>grabbing the data from teh objects
>

I think you're right. I do recall reading that it was for
overcoming atmospheric interference, but this was quite a while back.
I suppose that's what I get for posting about something from an
article I read around 10 years ago.
On the plus side, someone suggested looking up deconvolution,
and also that focus magic is a great program for addressing the
problem, so I ended up with some useful info even though I was
radically wrong about "adaptive optics".
It's a good group.