BBC - Lens does away with blurry snaps

Taken from http://news.bbc.co.uk/1/hi/technology/3643964.stm where you'll find
more...

Blurry images are caused by "lost information"
A specially shaped camera lens and processing method to ensure images are
always in focus has been developed.
Physicist Dr Andy Harvey said it was a "simple system with a simple lens"
which uses an optical encoder so that no information in images is lost.

Developed primarily for military night vision cameras, the technology could
find its way into camera phones.



--
Nick in Northallerton
www.whelan.me.uk
Also nickw7coc on
Yahoo Messenger
& on MSN Messenger
& www.skype.com !

"Nick" wrote in message
...
Taken from http://news.bbc.co.uk/1/hi/technology/3643964.stm where you'll
find
more...
Blurry images are caused by "lost information"
[snip]

This is not true. Although the lens and processing mentioned in your post
may well achieve less blurry results, it has nothing to do with lost
information. The blurriness itself contains information which allows one to
use holographic techniques to recover an unblurred image (though at greater
cost). That wouldn't be possible if the information was lost.
Grim

In message , Grimble Gromble
writes
"Nick" wrote in message
.. .
Taken from http://news.bbc.co.uk/1/hi/technology/3643964.stm where you'll
find
more...
Blurry images are caused by "lost information"
[snip]

This is not true.

It is a lot more nearly true than you have implied. In particular it is
true for any lens of finite aperture. Lack of measurements of the higher
spatial frequencies is essentially responsible for the point spread
function.

What they have done with wavefront coding may well be pretty clever, it
is hard to tell from the PR speak article on the BBC web page.

Although the lens and processing mentioned in your post
may well achieve less blurry results, it has nothing to do with lost
information. The blurriness itself contains information which allows one to
use holographic techniques to recover an unblurred image (though at greater
cost). That wouldn't be possible if the information was lost.
Grim

Anywhere that the sensor frequency response falls to zero you have
totally lost information. No amount of hand waving can get that back.
Tricks like requiring the sky brightness distribution to be everywhere
positive help enormously, but there are still ambiguities.

Regards,
--
Martin Brown

- And on Mon, 13 Sep 2004 17:31:40 GMT, it was spake thus in said in message "Grimble Gromble" :

"Nick" wrote in message
...
Taken from http://news.bbc.co.uk/1/hi/technology/3643964.stm where you'll
find
more...
Blurry images are caused by "lost information"
[snip]

This is not true. Although the lens and processing mentioned in your post
may well achieve less blurry results, it has nothing to do with lost
information. The blurriness itself contains information which allows one to
use holographic techniques to recover an unblurred image (though at greater
cost). That wouldn't be possible if the information was lost.
Grim


Did you read the whole of the BBC article...?
--
Nick in Northallerton
www.whelan.me.uk
Also nickw7coc on
Yahoo Messenger
& on MSN Messenger
& www.skype.com !

"Martin Brown" wrote in message
...

Anywhere that the sensor frequency response falls to zero you have
totally lost information. No amount of hand waving can get that back.
Tricks like requiring the sky brightness distribution to be everywhere
positive help enormously, but there are still ambiguities.


I can imagine! For example how could it differentiate between an out of
focus point object and a perfectly focussed "fuzzy blob" object without
making assumptions about what the object is supposed to look like

Robin

"Nick" wrote in message
...
[snip]
Did you read the whole of the BBC article...?

Yes.
Grim

"Robin Leadbeater" wrote in message
...
[snip]
I can imagine! For example how could it differentiate between an out of
focus point object and a perfectly focussed "fuzzy blob" object without
making assumptions about what the object is supposed to look like
Robin

How do you know whether you're looking at a point source or a fuzzy blob?
The assumption in the holographic technique is that the fuzzy blob you
typically sample is an imperfectly focussed point.
Grim

"Martin Brown" wrote in message
...
In message , Grimble Gromble
writes
"Nick" wrote in message
. ..
Taken from http://news.bbc.co.uk/1/hi/technology/3643964.stm where
you'll
find more...
Blurry images are caused by "lost information"
[snip]
This is not true.
It is a lot more nearly true than you have implied. In particular it is
true for any lens of finite aperture. Lack of measurements of the higher
spatial frequencies is essentially responsible for the point spread
function.

The implication of the article was that imperfect focus results in lost
information. Not so. The information is simply 'rearranged'.

What they have done with wavefront coding may well be pretty clever, it is
hard to tell from the PR speak article on the BBC web page.
[snip]
Anywhere that the sensor frequency response falls to zero you have totally
lost information. No amount of hand waving can get that back.
[snip]

I agree. But without knowing the details of the process, I can still
guarantee that they are not capturing more information than previously. The
information they record may well be different, but there isn't more of it.
Check out the 'no free lunch' principle.
Grim

"Grimble Gromble" wrote:
[snip]
I agree. But without knowing the details of the process, I can still
guarantee that they are not capturing more information than previously. The
information they record may well be different, but there isn't more of it.
Check out the 'no free lunch' principle.

One might gain some extra useable information with the infinite depth
of field.


Tim
--
Guns Don’t Kill People, Rappers Do.

In message , Grimble Gromble
writes
"Martin Brown" wrote in message
...
In message , Grimble Gromble
writes
"Nick" wrote in message
...
Taken from http://news.bbc.co.uk/1/hi/technology/3643964.stm where
you'll
find more...
Blurry images are caused by "lost information"
[snip]
This is not true.
It is a lot more nearly true than you have implied. In particular it is
true for any lens of finite aperture. Lack of measurements of the higher
spatial frequencies is essentially responsible for the point spread
function.

The implication of the article was that imperfect focus results in lost
information. Not so. The information is simply 'rearranged'.


Not so too :-) The high-frequency information is permanently lost, or it
could be recovered.