High resolution...through digital interpolation...

[Larry]

Archived from groups: rec.photo.digital (More info?)

In article <bteb51lnt3r0mrkkltemcpbg255093tiqv@4ax.com>, JPS@no.komm says...
> In message <MPG.1cbf4fe1d628e500989987@news.comcast.giganews.com>,
> Larry <lastingimagery@comcast.dotnet> wrote:
>
> >Actually, I just now used the Fuji to shoot some Crocus that just blossomed,
> >and I noticed that in all the shots where I used max apeture to limit depth
> >of field, the interpolated images
>
> All pixels in a regular bitmap are interpolated from that camera.
> --
>
> <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
> John P Sheehy <JPS@no.komm>
> ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
>
>

Yes they are. I was talking about the ones I let it interpolate to 12mp,
instead of the ones interpolated down to 6mp. I should have been clearer.



--
Larry Lynch
Mystic, Ct.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

On Thu, 07 Apr 2005 19:34:30 -0500, Ron Hunter <rphunter@charter.net>
wrote:

>John A. Stovall wrote:
>> On Thu, 07 Apr 2005 10:36:41 +0200, Mxsmanic <mxsmanic@hotmail.com>
>> wrote:
>>
>>
>>>Ron Hunter writes:
>>>
>>>
>>>>You make my point FOR me. In the early 1980's it was explained in great
>>>>detail, and with many engineers backing it, that nothing over 450 bps
>>>>was possible.
>>>
>>>By whom? Even ordinary voice communication is around 2400 bps with the
>>>simplest methods of modulation.
>>>
>>>
>>>>As we learn more about physics, more and more things become 'possible'.
>>>
>>>We have known for almost a hundred years what the bandwidth limits of
>>>ordinary telephone lines are. They have never changed.
>>>
>>>It is true that some types of modulation are limited to very low speeds.
>>>Higher speeds in modems came as newer types of modulation entered the
>>>scene. But there is an upper limit of 56 kbps in most cases, if for no
>>>other reason than the fact that so much traffic is digitized at 56 kbps
>>>today.
>>>
>>>
>>>>Before 1945 nuclear fission was just science fiction, and everyone knew
>>>>that was just something Jules Verne, and John W. Campbell, and his ilk
>>>>prattled on about, and was just about as likely as man ever being able
>>>>to go to the moon. Pure fantasy. Right!
>>>
>>>Actually, nuclear fission wasn't known at all prior to that. It was
>>>theorized, and immediately people saw the implications, but at first it
>>>wasn't clear whether explosive devices could be created using the
>>>principle. Until someone actually assembled a critical mass and let it
>>>run away, nobody was quite sure.
>>
>>
>> Actually nuclear fission know well before 1945 and Leo Szilard filed a
>> patten for a bomb in 1932.
>>
>> http/nuclearweaponarchive.org/Usa/Med/Discfiss.html
>>
>> Hahn observed in 1938 the first fission reactions in 1938.
>>
>>
>
>Non-selfsustaining reactions were known, but making a bomb from that was
>quite a step.

Just an engineering problem, no basic research or new insights
required.

>One could file a pattern for a faster than light spacecraft, but
>actually making one WORK is another matter. Little problem of power,
>for instance.

Once more you confuse engineering with the knowledge of basic
principles.


******************************************************************

"The past is foreign country: they do things differently there."


_The Go-Between_
L.P. Hartley
1895 - 1972

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

In article <lgd5e.7521$eF4.3896@fe03.lga>,
Ron Hunter <rphunter@charter.net> wrote:
>Philip Homburg wrote:
>> By early 1980's, 64 kbps connections had been in use for some time.
>>
>Make that mid 1980's, and they were way beyond the price range of home
>users at that time.

I should have said 50 kbps. The Arpanet started in 1969 and used 50 kbps
links.

The issue was whether it was possible at all, not whether it would be
affordable.


--
That was it. Done. The faulty Monk was turned out into the desert where it
could believe what it liked, including the idea that it had been hard done
by. It was allowed to keep its horse, since horses were so cheap to make.
-- Douglas Adams in Dirk Gently's Holistic Detective Agency

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

JPS@no.komm writes:

>All pixels in a regular bitmap are interpolated from that camera.

Are you sure? The simplest way to interpolate from a 6 MP rotated
raster to a 12 MP row/column raster is to keep all the original data
points and interpolate the new ones so they are equidistant between
the existing points in each row (or column, however you choose to look
at it). So you'd end up with 50% original pixels and 50% interpolated
pixels. This has got to be cheaper than generating 100% new pixels,
so I'd expect Fuji to do it.

On the other hand, if you select 6 MP output, pretty much all of those
pixels need to be interpolated because of the mismatch in sample pitch.

Dave

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Mxsmanic <mxsmanic@hotmail.com> wrote:

> Bubbabob writes:
>
>> I suggest that you look into the field of deconvolution algorithms.
>> Images CAN be improved.
>
> No, they cannot.
>
>> NASA/HST do it every day. Not to mention the NRO and a few
>> other black ops.
>
> No, they don't. It's mathematically impossible, even for the spooks.
>

Nonsense. Deconvolution can compensate for an imperfect optical system
and bring out of focus areas back into focus (within limits). I can
generate about 2 stops worth of additional depth of field in many of my
images by the proper use of deconvolution. As long as you don't cross the
line (and you'll know when you do) it provides sharpening with very few
of the disagreeable artifacts of USM. The downside is that it's about 30
times slower.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Owamanga <owamanga(not-this-bit)@hotmail.com> writes:

> The fiber is the same fiber. We've laid new stuff, but the old stuff
> hasn't been retired.

Fiber in general isn't that old.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Ron Hunter writes:

> I should be so lucky as to have fiber! The telephone lines here are
> single strand copper wrapped in PAPER wrapping, and inside a LEAD tube.

I don't know about the U.S., but in France, practically everything is
fiber, except the last mile, or so I've been told.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Mxsmanic wrote:
> Ron Hunter writes:
>
>
>>I should be so lucky as to have fiber! The telephone lines here are
>>single strand copper wrapped in PAPER wrapping, and inside a LEAD tube.
>
>
> I don't know about the U.S., but in France, practically everything is
> fiber, except the last mile, or so I've been told.
>
If so, then there isn't much DSL service there, right?


--
Ron Hunter rphunter@charter.net

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Ron Hunter writes:

> If so, then there isn't much DSL service there, right?

DSL is very widespread in France. Why would fiber be an obstacle to
that?

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Bubbabob writes:

> Nonsense. Deconvolution can compensate for an imperfect optical system
> and bring out of focus areas back into focus (within limits). I can
> generate about 2 stops worth of additional depth of field in many of my
> images by the proper use of deconvolution.

If you don't mind extremely noisy images.

Nothing you do can add any detail that wasn't present in the original
image. Blurring can only be undone if you can make safe assumptions
about the original scene, and those assumptions are completely safe only
if you know the content of the original scene, and if you know the
content of the original scene, you don't need to undo the blurring.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

On Fri, 08 Apr 2005 06:04:03 +0200, Mxsmanic <mxsmanic@hotmail.com>
wrote:

>Owamanga <owamanga(not-this-bit)@hotmail.com> writes:
>
>> The fiber is the same fiber. We've laid new stuff, but the old stuff
>> hasn't been retired.
>
>Fiber in general isn't that old.

First large scale commercial telephone use in 1982, first
transatlantic submarine cable around 1988. First transoceanic
self-healing fiber ring, 1996.

DSL was invented by Bell & AT&T in 1988, 8 years after fiber was first
put to use, but wide scale deployment came *way* after that. My point
is for large parts of the US and other advanced countries, the
technology and backbone infrastructure was there already waiting for
DSL, not put in because of DSL. In fact, the largest delay was simply
waiting for the internet to become popular.

To put this into perspective, it was ten years after DSL, late 1998
when V.90 was ratified, to provide modems with 56kbps downstream,
33kbps upstream. K56flex development started in 1996, and equipment to
support it was rolling out to the big ISPs that same year. All of this
happened 10-15 years *after* fiber.

--
Owamanga!
http/www.pbase.com/owamanga

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Mxsmanic wrote:
>
> Nothing you do can add any detail that wasn't present in the original
> image. Blurring can only be undone if you can make safe assumptions
> about the original scene, and those assumptions are completely safe
only
> if you know the content of the original scene, and if you know the
> content of the original scene, you don't need to undo the blurring.
>
You are not totally correct here.
First it is important to realize just what the effect of blurring is,
this is simply a reduction of the higher spatial frequencies in the
image, a low pass filter if you will. These frequencies can be
selectively boosted by an appropriate high pass filter. There is a
limit to this due to the presents of noise, which will be boosted along
with the image detail but the high pass filter. How much de-blurring
you can do depends on how good of a signal to noise ratio you have in
the image.

So you are not adding detail that was not there, you are making detail
visible that was not before because it was to low in amplitude due to
the blurring.

There was a whole lot of work done in the 70s on doing this using
coherent optics, for those that might remember the 70s it was not so
easy to digitize an image and run a digital filter on it.

Scott

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Mxsmanic <mxsmanic@hotmail.com> wrote:

> Bubbabob writes:
>
>> Nonsense. Deconvolution can compensate for an imperfect optical system
>> and bring out of focus areas back into focus (within limits). I can
>> generate about 2 stops worth of additional depth of field in many of my
>> images by the proper use of deconvolution.
>
> If you don't mind extremely noisy images.

My digital prints have less noise than my 120 negs have grain.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Mxsmanic wrote:
> Ron Hunter writes:
>
>
>>If so, then there isn't much DSL service there, right?
>
>
> DSL is very widespread in France. Why would fiber be an obstacle to
> that?
>
Because DSL works ONLY on copper connections. It is possible to place
DSLAMs at the fiber termination, but this increases the infrastructure cost.


--
Ron Hunter rphunter@charter.net

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Mxsmanic wrote:
> Bubbabob writes:
>
>
>>Nonsense. Deconvolution can compensate for an imperfect optical system
>>and bring out of focus areas back into focus (within limits). I can
>>generate about 2 stops worth of additional depth of field in many of my
>>images by the proper use of deconvolution.
>
>
> If you don't mind extremely noisy images.
>
> Nothing you do can add any detail that wasn't present in the original
> image. Blurring can only be undone if you can make safe assumptions
> about the original scene, and those assumptions are completely safe only
> if you know the content of the original scene, and if you know the
> content of the original scene, you don't need to undo the blurring.
>
Just plain NOT TRUE. There are some excellent programs that will
minimize motion blur by analyzing the direction of the blur, and working
to correct it. The process requires a lot of processing and the results
are much less than perfect, but the process IS possible.


--
Ron Hunter rphunter@charter.net

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Owamanga <owamanga(not-this-bit)@hotmail.com> writes:

> All of this happened 10-15 years *after* fiber.

And flat-panel displays predate this by 20 years. Microwave ovens
predate it by 40 years at least, as does digital voice transmission.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

In message <d354ae$ddf$1@mughi.cs.ubc.ca>,
davem@cs.ubc.ca (Dave Martindale) wrote:

>JPS@no.komm writes:
>
>>All pixels in a regular bitmap are interpolated from that camera.
>
>Are you sure?

No; as a matter of fact, I really meant something like, "All images with
a regular bitmap ..."

> The simplest way to interpolate from a 6 MP rotated
>raster to a 12 MP row/column raster is to keep all the original data
>points and interpolate the new ones so they are equidistant between
>the existing points in each row (or column, however you choose to look
>at it). So you'd end up with 50% original pixels and 50% interpolated
>pixels. This has got to be cheaper than generating 100% new pixels,
>so I'd expect Fuji to do it.


--

<>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
John P Sheehy <JPS@no.komm>
><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Ron Hunter writes:

> Because DSL works ONLY on copper connections.

I excluded the last mile (the local subscriber loop), as you may recall.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Ron Hunter writes:

> Just plain NOT TRUE. There are some excellent programs that will
> minimize motion blur by analyzing the direction of the blur, and working
> to correct it. The process requires a lot of processing and the results
> are much less than perfect, but the process IS possible.

None of these programs can add detail that wasn't originally there.

--
Transpose hotmail and mxsmanic in my e-mail address to reach me directly.

 

[Guest]

Archived from groups: rec.photo.digital (More info?)

Mxsmanic wrote:
> Ron Hunter writes:
>
>
>>Just plain NOT TRUE. There are some excellent programs that will
>>minimize motion blur by analyzing the direction of the blur, and working
>>to correct it. The process requires a lot of processing and the results
>>are much less than perfect, but the process IS possible.
>
>
> None of these programs can add detail that wasn't originally there.
>
Quibbling. If an interpolated pixel is the same color, and in the same
position as a 'real' pixel would be in a camera with a finer resolution,
what is the difference? If 50% of the pixels are in the same place as
the real pixels would be, then the picture would look a LOT better, so
does it matter HOW they got to your screen?

Once the pixel is in the file, and displayed on your screen, can you
pick out all the interpolated ones?


--
Ron Hunter rphunter@charter.net