Re: Cameras Create Highly Revealing Snapshots

Discussion in 'Digital Photography' started by hanson, Jul 2, 2011.

  1. hanson

    hanson Guest

    Cool, Sam,
    Could someone post how the best el-pixel resolution
    compares to the best resolution that was achieved
    with the old Silver halide emulsions?
    Which one gives finer details?
    >
    >

    "Sam Wormley" <> wrote in message
    news:...
    > Cameras Create Highly Revealing Snapshots
    >
    > http://www.scientificamerican.com/article.cfm?id=gigapixel-camera-revealed
    >
    > "Through its Advanced Wide Field of View Architectures for Image
    > Reconstruction and Exploitation program, DARPA has for the past year been
    > working on ways to develop a camera that can take a gigapixel-quality
    > image in a single snapshot. This approach is novel, given that today's
    > gigapixel images actually consist of several megapixel-sized images pieced
    > together digitally to provide a high level of detail over a large area.
    > This is often done using a long-lens digital single-lens reflex (SLR)
    > camera placed atop a motorized mount. Software controls the movement of
    > the camera, which captures a mosaic of hundreds or even thousands of
    > images that, when placed together, create a single, high-resolution scene
    > that maintains its clarity even when the viewer zooms in on a specific
    > area. DARPA plans to invest $25 million over a three-and-a-half-year
    > period in its program, which includes a component called Maximally
    > scalable Optical Sensor Array Imaging with Computation (MOSAIC)".
    >
    > See:
    > http://www.scientificamerican.com/article.cfm?id=gigapixel-camera-revealed
     
    hanson, Jul 2, 2011
    #1
    1. Advertising

  2. hanson

    hanson Guest

    "Paul Furman" <> wrote:
    - hanson wrote:
    -- "Sam Wormley" <> wrote:
    >
    >

    hanson wrote:
    Well, that may be so, Paul, but what I posted and
    wanted to hear was and is:
    Could someone post how the best el-pixel resolution
    compares to the best resolution that was achieved
    with the old Silver halide emulsions?
    Which one gives finer details?
     
    hanson, Jul 3, 2011
    #2
    1. Advertising

  3. hanson

    hanson Guest

    Cool, Paul!... Thanks!
    >

    "Paul Furman" <> wrote in message
    news:...
    > hanson wrote:
    >> Paul Furman wrote:
    >> - hanson wrote:
    >> -- Sam Wormley wrote:
    >>>

    >> hanson wrote:
    >> Well, that may be so, Paul, but what I posted and
    >> wanted to hear was and is:
    >> Could someone post how the best el-pixel resolution
    >> compares to the best resolution that was achieved
    >> with the old Silver halide emulsions?
    >> Which one gives finer details?

    >

    Paul wrote:
    > I missed the O.P. so replied there.
    > The article did mention large format film as one option, although I'd
    > guess color is helpful for many of their spy satellite needs.
    >
    > But I'm sure black and white film can get a heck of a lot more detail than
    > any large format sensor, because there are no large format sensors
    > available other than scanning backs. Another concept in the article,
    > shifted a smaller sensor (35mm square) around to catch a mosaic but that's
    > not a single snapshot; it still takes time to cycle through.
    >
    > Still, I'm pretty sure black and white film exceeds digital for dynamic
    > range plus resolution. Not true for color. In fact, if you use a
    > monochrome sensor (scientific grade as used in their prototypes), that may
    > get the advantage back.
    >
    > Another thing the article mentioned was using an older film lens because
    > there are no large format lenses optimized for digital. Even though it's
    > not super-high resolution, they determined that the larger format captured
    > more detail than medium format. So it's not super-high resolution in terms
    > of lp/mm.
    >
    > Also, from the comments:
    > "As someone who worked with satellite pictures during the Vietnam War, I
    > find this technology amazing. We had to launch very heavy, low-earth-orbit
    > satellites from Vandenberg AFB, and the film was returned to earth in
    > capsules that were captured in mid-air over the South Pacific. The
    > resolution was breathtaking.
    > This new methodology, even with its drawbacks, sounds far faster, and as
    > for cost, I do not imagine it is possible for it to be as expensive as a
    > Titan liquid-fuel rocket with multiple solid-fuel strap-on boosters."
    >
    >
    >> Paul Furman wrote:
    >>>> Sam Wormley wrote
    >>>>> Cameras Create Highly Revealing Snapshots
    >>>>> http://www.scientificamerican.com/article.cfm?id=gigapixel-camera-revealed

    >

    Paul wrote:
    > OH, here, on page two:
    > "Another large-format approach to taking gigapixel snapshots is the
    > Gigapixl Project, which physicist Graham Flint formed about a decade ago.
    > Gigapixl's camera uses 23-by-46-centimeter film—the same used in military
    > spy planes such as the U-2, to capture images—which is then scanned and
    > digitized to create images up to four gigapixels in size."
    >
    >>>>> "Through its Advanced Wide Field of View Architectures for Image
    >>>>> Reconstruction and Exploitation program, DARPA has for the past year
    >>>>> been working on ways to develop a camera that can take a
    >>>>> gigapixel-quality image in a single snapshot. This approach is novel,
    >>>>> given that today's gigapixel images actually consist of several
    >>>>> megapixel-sized images pieced together digitally to provide a high
    >>>>> level of detail over a large area. This is often done using a
    >>>>> long-lens digital single-lens reflex (SLR) camera placed atop a
    >>>>> motorized mount. Software controls the movement of the camera, which
    >>>>> captures a mosaic of hundreds or even thousands of images that, when
    >>>>> placed together, create a single, high-resolution scene that maintains
    >>>>> its clarity even when the viewer zooms in on a specific area. DARPA
    >>>>> plans to invest $25 million over a three-and-a-half-year period in its
    >>>>> program, which includes a component called Maximally scalable Optical
    >>>>> Sensor Array Imaging with Computation (MOSAIC)".
    >>>
    >>> This is the most interesting of the options they mention:
    >>> http://www.scientificamerican.com/s...&photo_id=4E06507E-C5F7-290F-4ADE7A53BE3A83BC
    >>> -lens is a sphere with microlenses on the back side which project onto a
    >>> hemisphere of small sensors around the lens-ball.
    >>>

    Paul wrote:
    >>> I never heard of this idea that a sphere shaped single element lens had
    >>> optical advantages of fewer aberrations though. Maybe the problem is it
    >>> won't focus on a flat sensor/film so this solves that problem.
     
    hanson, Jul 3, 2011
    #3
  4. On Sat, 02 Jul 2011 17:27:51 -0700, Paul Furman <>
    wrote:

    >hanson wrote:
    >> Paul Furman wrote:
    >> - hanson wrote:
    >> -- Sam Wormley wrote:
    >>>
    >>>

    >> hanson wrote:
    >> Well, that may be so, Paul, but what I posted and
    >> wanted to hear was and is:
    >> Could someone post how the best el-pixel resolution
    >> compares to the best resolution that was achieved
    >> with the old Silver halide emulsions?
    >> Which one gives finer details?

    >
    >I missed the O.P. so replied there.
    >
    >The article did mention large format film as one option, although I'd
    >guess color is helpful for many of their spy satellite needs.
    >
    >But I'm sure black and white film can get a heck of a lot more detail
    >than any large format sensor, because there are no large format sensors
    >available other than scanning backs. Another concept in the article,
    >shifted a smaller sensor (35mm square) around to catch a mosaic but
    >that's not a single snapshot; it still takes time to cycle through.
    >
    >Still, I'm pretty sure black and white film exceeds digital for dynamic
    >range plus resolution. Not true for color. In fact, if you use a
    >monochrome sensor (scientific grade as used in their prototypes), that
    >may get the advantage back.
    >
    >Another thing the article mentioned was using an older film lens because
    >there are no large format lenses optimized for digital. Even though it's
    >not super-high resolution, they determined that the larger format
    >captured more detail than medium format. So it's not super-high
    >resolution in terms of lp/mm.
    >
    >Also, from the comments:
    >"As someone who worked with satellite pictures during the Vietnam War, I
    >find this technology amazing. We had to launch very heavy,
    >low-earth-orbit satellites from Vandenberg AFB, and the film was
    >returned to earth in capsules that were captured in mid-air over the
    >South Pacific. The resolution was breathtaking.
    >This new methodology, even with its drawbacks, sounds far faster, and as
    >for cost, I do not imagine it is possible for it to be as expensive as a
    >Titan liquid-fuel rocket with multiple solid-fuel strap-on boosters."
    >......


    With a consumer type 24x36 camera the resolution limiting factor
    is the optics, not the photographic film.
    Diffraction limits such a photo to about 20 megapixels in theory.

    Usually it is much less, about 4 Megapixels with a "good" lens.
    Even less with el-cheapo lenses and wrong setting.
    Any photographic optic I put on my camera has it's optimal
    resolution at a certain aperture only. Put superglue on the
    aperture ring, and never use zoom lenses.

    Average foto 24x36 slides have less than 1 megapixel,
    and that was considered a good foto before the digital age.


    w.
     
    Helmut Wabnig, Jul 3, 2011
    #4
  5. hanson

    RichA Guest

    On Jul 3, 8:13 am, Neil Ellwood <>
    wrote:
    > On Sat, 02 Jul 2011 11:43:44 -0700, hanson wrote:
    > > Cool, Sam,
    > > Could someone post how the best el-pixel resolution compares to the best
    > > resolution that was achieved with the old Silver halide emulsions? Which
    > > one gives finer details?

    >
    > > "Sam Wormley" <> wrote in message
    > >news:...
    > >> Cameras Create Highly Revealing Snapshots

    >
    > >>http://www.scientificamerican.com/article.cfm?id=gigapixel-camera-

    > revealed
    >
    > >> "Through its Advanced Wide Field of View Architectures for Image
    > >> Reconstruction and Exploitation program, DARPA has for the past year
    > >> been working on ways to develop a camera that can take a
    > >> gigapixel-quality image in a single snapshot. This approach is novel,
    > >> given that today's gigapixel images actually consist of several
    > >> megapixel-sized images pieced together digitally to provide a high
    > >> level of detail over a large area. This is often done using a long-lens
    > >> digital single-lens reflex (SLR) camera placed atop a motorized mount.
    > >> Software controls the movement of the camera, which captures a mosaic
    > >> of hundreds or even thousands of images that, when placed together,
    > >> create a single, high-resolution scene that maintains its clarity even
    > >> when the viewer zooms in on a specific area. DARPA plans to invest $25
    > >> million over a three-and-a-half-year period in its  program, which
    > >> includes a component called Maximally scalable Optical Sensor Array
    > >> Imaging with Computation (MOSAIC)".

    >
    > >> See:
    > >>http://www.scientificamerican.com/article.cfm?id=gigapixel-camera-

    >
    > revealed
    >
    > Believe it or not but it is the result I am interested in. The final
    > picture is my aim.
    >
    > --
    > Neil
    > Linux counter 335851
    > delete ‘l’ and reverse ‘r’ and’a’


    Whatever the resolution of the film, divide it by four by the time it
    hits the print, no matter what lenses or techniques are used.
     
    RichA, Jul 3, 2011
    #5
  6. Helmut Wabnig:

    > Usually it is much less, about 4 Megapixels with a
    > "good" lens. Even less with el-cheapo lenses and
    > wrong setting. Any photographic optic I put on my
    > camera has it's optimal
    > resolution at a certain aperture only. Put super-
    > glue on the aperture ring, and never use zoom
    > lenses. P Average foto 24x36 slides have less
    > than 1 megapixel, and that was considered a good
    > foto before the digital age.


    I must be serously misunderstanding something,
    because both figures seem to me severely underesti-
    mated. Here is a fragment of a 2900-dpi scan of an
    old Soviet slide, shot with a consumer-grade camera
    through a Helios lens, well known for its softness.
    It is much more than one or four megapixels:

    http://xmages.net/storage/10/1/0/6/8/upload/d116ae3b.jpg

    With modern 50-ISO B/W film, the resolution of my
    2900-dpi scanner is just not enough, so twelve
    megapixels is way below the practical achivable
    limit.

    Anton
     
    Anton Shepelev, Jul 4, 2011
    #6
  7. On Tue, 5 Jul 2011 00:15:51 +0400, Anton Shepelev
    <> wrote:

    >Helmut Wabnig:
    >
    >> Usually it is much less, about 4 Megapixels with a
    >> "good" lens. Even less with el-cheapo lenses and
    >> wrong setting. Any photographic optic I put on my
    >> camera has it's optimal
    >> resolution at a certain aperture only. Put super-
    >> glue on the aperture ring, and never use zoom
    >> lenses. P Average foto 24x36 slides have less
    >> than 1 megapixel, and that was considered a good
    >> foto before the digital age.

    >
    >I must be serously misunderstanding something,
    >because both figures seem to me severely underesti-
    >mated. Here is a fragment of a 2900-dpi scan of an
    >old Soviet slide, shot with a consumer-grade camera
    >through a Helios lens, well known for its softness.
    >It is much more than one or four megapixels:
    >
    > http://xmages.net/storage/10/1/0/6/8/upload/d116ae3b.jpg
    >
    >With modern 50-ISO B/W film, the resolution of my
    >2900-dpi scanner is just not enough, so twelve
    >megapixels is way below the practical achivable
    >limit.
    >
    >Anton



    Downscaled from 3,21 Mb to 384 kByte.
    how much of a difference do you see?

    http://img827.imageshack.us/img827/8982/comparisonc.jpg

    A higer resolution scanner will only produce more grain noise.
    The image is already overblown and the countours cannot be
    made sharper and you cannot get more details out of the foto.

    "Empty magnification" is produced by the scanner.
    Set the scanner resolution down until you notice some details
    are missing, e.g. the white lines on the red kilt cloth blend.

    The smallest discernible detail measured in line-pairs per millimeter
    defines resolution, not the film emulsion grain speckles which may be
    much smaller than any photographed detail.

    w.
     
    Helmut Wabnig, Jul 4, 2011
    #7
  8. hanson

    Androcles Guest

    "Helmut Wabnig" <hwabnig@.- --- -.dotat> wrote in message
    news:eek:...
    | On Tue, 5 Jul 2011 00:15:51 +0400, Anton Shepelev
    | <> wrote:
    |
    | >Helmut Wabnig:
    | >
    | >> Usually it is much less, about 4 Megapixels with a
    | >> "good" lens. Even less with el-cheapo lenses and
    | >> wrong setting. Any photographic optic I put on my
    | >> camera has it's optimal
    | >> resolution at a certain aperture only. Put super-
    | >> glue on the aperture ring, and never use zoom
    | >> lenses. P Average foto 24x36 slides have less
    | >> than 1 megapixel, and that was considered a good
    | >> foto before the digital age.
    | >
    | >I must be serously misunderstanding something,
    | >because both figures seem to me severely underesti-
    | >mated. Here is a fragment of a 2900-dpi scan of an
    | >old Soviet slide, shot with a consumer-grade camera
    | >through a Helios lens, well known for its softness.
    | >It is much more than one or four megapixels:
    | >
    | > http://xmages.net/storage/10/1/0/6/8/upload/d116ae3b.jpg
    | >
    | >With modern 50-ISO B/W film, the resolution of my
    | >2900-dpi scanner is just not enough, so twelve
    | >megapixels is way below the practical achivable
    | >limit.
    | >
    | >Anton
    |
    |
    | Downscaled from 3,21 Mb to 384 kByte.

    B/W film existed long before the digital age, you moron.
    It's been around since the mid-1800s. You don't need a
    darkroom for a digital camera, why are you posting your
    babble to rec.photo.darkroom, wabnigger?
     
    Androcles, Jul 4, 2011
    #8
  9. Helmut Wabnig:

    > Downscaled from 3,21 Mb to 384 kByte.
    > how much of a difference do you see?


    I can't figure what you have done. The IrfanView
    screenshot says that the compressed image size is
    different but the real size, which determines the
    resolution, is the same (11.37 Mb).

    Let's operate linear (pixels per inch) resulution
    instead of the quadratic one (image size, megapix-
    els, e.t.c.), which is misleading.

    I have downsized the original fragment by a factor
    of two:

    http://xmages.net/storage/10/1/0/1/5/upload/8017b14d.png

    And here's a side-by-side screenshot comparing it
    with the original fragment. The downscaled verstion
    have been resampled by the same factor of two, with-
    out interpolation, which is OK for integer factors.

    http://xmages.net/storage/10/1/0/d/0/upload/ea5735f2.png

    Notice the loss of detail in the left image. Here's
    the same comparison with interpolation:

    http://xmages.net/storage/10/1/0/0/8/upload/b6854364.png

    Still less sharp than the original.

    From this I conclude that there is more than
    2900/2=1450 dpi resolution in this particular slide.

    > A higer resolution scanner will only produce more
    > grain noise.


    A higher resolution will only improve the look of
    the grain, because it will be done above the charac-
    teristic frequency of the grain:

    http://www.photoscientia.co.uk/Grain.htm

    > "Empty magnification" is produced by the scanner.
    > Set the scanner resolution down until you notice
    > some details
    > are missing, e.g. the white lines on the red kilt
    > cloth blend.


    This is bad advice when applied to scanning film. In
    addition to the above, the level of details in not a
    binary thing. It is also incorrect to judge it only
    by high-contrast areas. Pay attention to medium- and
    low-contrast areas as well -- the grey cap for exam-
    ple.

    > The smallest discernible detail measured in line-
    > pairs per millimeter defines resolution, not the
    > film emulsion grain speckles which may be much
    > smaller than any photographed detail.


    Yes, but this measurement can only be used to com-
    pare against other measurement in the same condi-
    tions -- lightning, chart contrast, e.t.c.

    There is no well-defined boundary between meaningful
    detail and grain. As the frequency increases, the
    ratio of detail vs. grain noise decreases, and with
    good film scanned at 4000 dpi it still far from
    zero.

    Anton
     
    Anton Shepelev, Jul 5, 2011
    #9
  10. On Mon, 4 Jul 2011 22:48:44 +0100, "Androcles"
    <.2011> wrote:

    >
    >"Helmut Wabnig" <hwabnig@.- --- -.dotat> wrote in message
    >news:eek:...
    >| On Tue, 5 Jul 2011 00:15:51 +0400, Anton Shepelev
    >| <> wrote:
    >|
    >| >Helmut Wabnig:
    >| >
    >| >> Usually it is much less, about 4 Megapixels with a
    >| >> "good" lens. Even less with el-cheapo lenses and
    >| >> wrong setting. Any photographic optic I put on my
    >| >> camera has it's optimal
    >| >> resolution at a certain aperture only. Put super-
    >| >> glue on the aperture ring, and never use zoom
    >| >> lenses. P Average foto 24x36 slides have less
    >| >> than 1 megapixel, and that was considered a good
    >| >> foto before the digital age.
    >| >
    >| >I must be serously misunderstanding something,
    >| >because both figures seem to me severely underesti-
    >| >mated. Here is a fragment of a 2900-dpi scan of an
    >| >old Soviet slide, shot with a consumer-grade camera
    >| >through a Helios lens, well known for its softness.
    >| >It is much more than one or four megapixels:
    >| >
    >| > http://xmages.net/storage/10/1/0/6/8/upload/d116ae3b.jpg
    >| >
    >| >With modern 50-ISO B/W film, the resolution of my
    >| >2900-dpi scanner is just not enough, so twelve
    >| >megapixels is way below the practical achivable
    >| >limit.
    >| >
    >| >Anton
    >|
    >|
    >| Downscaled from 3,21 Mb to 384 kByte.
    >
    >B/W film existed long before the digital age, you moron.
    >It's been around since the mid-1800s. You don't need a
    >darkroom for a digital camera, why are you posting your
    >babble to rec.photo.darkroom, wabnigger?
    >


    Especially for you, Andro,
    because those are the only questions
    which you are capable of asking.


    w.
     
    Helmut Wabnig, Jul 5, 2011
    #10
  11. On Tue, 5 Jul 2011 12:30:38 +0400, Anton Shepelev
    <anton.txt@g{oogle}mail.com> wrote:

    >...........
    >Yes, but this measurement can only be used to com-
    >pare against other measurement in the same condi-
    >tions -- lightning, chart contrast, e.t.c.



    The resolution of a perfect fotografic objective is given by
    the size of the Airy disk.
    That requires a perfectly made lens, with all the consumer
    and professional photography lenses you are far from the
    theoretically possible diffration limit, because a lot of other
    optical errors are summed, and in total are much larger
    than the diffraction allows.
    The smaller the aperture, the less resolution you get.

    You obviously do not understand what optical resolution
    means in terms of optics.

    On your fotographic scans the sharpest details come from the dust
    particles which accumulate on the emulsion. You have no chance to
    avoid that, except when working in a special clean room.
    Compare the dust grains with your sharpest fotografic details.
    I leave you alone now with your stubborness.

    Ask Androcles.

    w.
     
    Helmut Wabnig, Jul 5, 2011
    #11
  12. hanson

    Androcles Guest

    "Helmut Wabnig" <hwabnig@.- --- -.dotat> wrote in message
    news:...
    | On Tue, 5 Jul 2011 12:30:38 +0400, Anton Shepelev
    | <anton.txt@g{oogle}mail.com> wrote:
    |
    | >...........
    | >Yes, but this measurement can only be used to com-
    | >pare against other measurement in the same condi-
    | >tions -- lightning, chart contrast, e.t.c.
    |
    |
    | The resolution of a perfect fotografic objective is given by
    | the size of the Airy disk.
    | That requires a perfectly made lens, with all the consumer
    | and professional photography lenses you are far from the
    | theoretically possible diffration limit, because a lot of other
    | optical errors are summed, and in total are much larger
    | than the diffraction allows.
    | The smaller the aperture, the less resolution you get.
    |
    | You obviously do not understand what optical resolution
    | means in terms of optics.
    |
    | On your fotographic scans the sharpest details come from the dust
    | particles which accumulate on the emulsion. You have no chance to
    | avoid that, except when working in a special clean room.
    | Compare the dust grains with your sharpest fotografic details.
    | I leave you alone now with your stubborness.
    |
    | Ask Androcles.
    |
    Run away, wabnigger. Make sure your photographically made
    pixel array didn't have any dust on it.
     
    Androcles, Jul 5, 2011
    #12
  13. hanson

    hanson Guest

    "Helmut Wabnig" <hwabnig@.- --- -.dotat> wrote in message
    news:...
    > On Tue, 5 Jul 2011 12:30:38 +0400, Anton Shepelev
    > <anton.txt@g{oogle}mail.com> wrote:
    >
    >>...........

    >

    Anton wrote:
    >>Yes, but this measurement can only be used to com-
    >>pare against other measurement in the same condi-
    >>tions -- lightning, chart contrast, e.t.c.

    >
    >

    Wabie wrote:
    > The resolution of a perfect fotografic objective is given by
    > the size of the Airy disk.
    > That requires a perfectly made lens, with all the consumer
    > and professional photography lenses you are far from the
    > theoretically possible diffration limit, because a lot of other
    > optical errors are summed, and in total are much larger
    > than the diffraction allows.
    > The smaller the aperture, the less resolution you get.
    >
    > You obviously do not understand what optical resolution
    > means in terms of optics.
    >
    > On your fotographic scans the sharpest details come from the dust
    > particles which accumulate on the emulsion. You have no chance to
    > avoid that, except when working in a special clean room.
    > Compare the dust grains with your sharpest fotografic details.
    > I leave you alone now with your stubborness.
    >
    >....

    .....
    ....
    hanson wrote:
    Not so fast Wabie. There may be some linguistic issues
    in your last para. "sharpest details come from the dust
    particles which accumulate on the emulsion."... Could
    you rephrase that, so that the meaning becomes clearer
     
    hanson, Jul 5, 2011
    #13
  14. Helmut Wabnig:

    > That requires a perfectly made lens, with all the
    > consumer and professional photography lenses you
    > are far from the theoretically possible diffration
    > limit, because a lot of other optical errors are
    > summed, and in total are much larger than the
    > diffraction allows.


    Which means the theory you are referring to, without
    providing the derivation of the highest theoretical
    limit of resolution, can not be an argument, and the
    error can only be accounted for by real measurements
    of the resulting negative.

    > You obviously do not understand what optical reso-
    > lution means in terms of optics.


    I did not even refer to this. After you faked the
    downscaling of my example scan of a low-quality
    slide, I did it myself and thus experimentally dis-
    proved your statement about the ridiculously low
    resolution of slides. Did you ignore that part of my
    post for having nothing to say?

    What about you? Do you realize how to convert con-
    vert lines per inch into digital pixels per inch?

    > On your fotographic scans the sharpest details
    > come from the dust particles which accumulate on
    > the emulsion.
    > [...]
    > Compare the dust grains with your sharpest
    > fotografic details.


    So what of it? Dust will always be sharper than the
    image on the negative. What's the point of stating
    this obvious fact? It has nothing to do with the
    resolution of the image.

    > I leave you alone now with your stubborness.


    You are suspiciously quick to do it.

    Anton
     
    Anton Shepelev, Jul 6, 2011
    #14
  15. hanson

    Martin Brown Guest

    On 06/07/2011 08:57, Anton Shepelev wrote:
    > Helmut Wabnig:
    >
    >> That requires a perfectly made lens, with all the
    >> consumer and professional photography lenses you
    >> are far from the theoretically possible diffration
    >> limit, because a lot of other optical errors are
    >> summed, and in total are much larger than the
    >> diffraction allows.

    >
    > Which means the theory you are referring to, without
    > providing the derivation of the highest theoretical
    > limit of resolution, can not be an argument, and the
    > error can only be accounted for by real measurements
    > of the resulting negative.


    The Rayleigh criterion 1.22lambda/D (D= diameter) holds for determining
    the effective maximum angular resolution of any circular aperture.

    There is nothing useful on the film that can be recorded at any higher
    spatial frequencies than the working aperture of the lens can capture.

    And you can put an upper bound on it based on the wavelength of light
    used and the working aperture by assuming perfect diffraction limited
    optics (usually only true for f5 or slower). At wider apertures the
    image is limited by various optical aberrations.

    Taking 550nm as mid green visible light the diffraction limited bound
    works out to be 0.7f um or if you prefer 1400/f lp/mm.

    Where f is the focal ratio of the lens. Apart from a handful of pro
    lenses that are truly diffraction limited at f4 this puts a lower bound
    on the FWHM of 3.5 um or 280 lp/mm so a 36mm frame needs at most 9M
    samples assuming a perfect recording medium. Realistically most lenses
    are a factor of two worse than this rough estimate in practice.
    >
    >> You obviously do not understand what optical reso-
    >> lution means in terms of optics.

    >
    > I did not even refer to this. After you faked the
    > downscaling of my example scan of a low-quality
    > slide, I did it myself and thus experimentally dis-
    > proved your statement about the ridiculously low
    > resolution of slides. Did you ignore that part of my
    > post for having nothing to say?


    Your low quality slide at high resolution consists of mostly grain noise
    and the image itself is very soft. He hasn't lost much by downsampling
    it - mostly edge detail around the dust.
    >
    > What about you? Do you realize how to convert con-
    > vert lines per inch into digital pixels per inch?


    Even the best 50ASA slide films like Fuji Velvia have an MTF that pretty
    much goes rapidly to zero above 100lp/mm eg.

    http://www.fujifilm.co.uk/professional/films/pdfs/fujichrome_velvia50.pdf

    It's MTF is down to 50% at 50lp/mm and they make no claims at all in the
    datsheet for anything above 70. Their 100 ASA slide tails off quicker
    still and is down to 33% at 50lp/mm.
    >
    >> On your fotographic scans the sharpest details
    >> come from the dust particles which accumulate on
    >> the emulsion.
    >> [...]
    >> Compare the dust grains with your sharpest
    >> fotografic details.

    >
    > So what of it? Dust will always be sharper than the
    > image on the negative. What's the point of stating
    > this obvious fact? It has nothing to do with the
    > resolution of the image.


    Actually it does - there isn't a lot of point in carefully digitising
    the exact shape of every piece of dust and grain boundary on a soft
    focus image like that. The high frequency grain noise could be separated
    from the true image with a bit of cunning adaptive filtering.

    There is some merit in slight (eg 1.5x) oversampling of images but the
    current generation of digital sensors are now close enough to the
    information theoretic sampling that they seldom limit image quality.

    Regards,
    Martin Brown
     
    Martin Brown, Jul 6, 2011
    #15
  16. Thanks for the reply, Martin.

    > Apart from a handful of pro lenses that are truly
    > diffraction limited at f4 this puts a lower bound
    > on the FWHM of 3.5 um or 280 lp/mm so a 36mm frame
    > needs at most 9M samples assuming a perfect
    > recording medium. Realistically most lenses are a
    > factor of two worse than this rough estimate in
    > practice.


    9M is about 100 pixels/mm. How can 280 line pairs
    fit into 100 pixels, if you do consider a digital
    image a medium close to perfect?

    Also, if two systems connected is series have corre-
    sponding resolutions Res1 and Res2, then the final
    resolution is roughly:

    1/Res = 1/Res1 + 1/Res2

    Doesn't it mean that the scanning resolution should
    exceed that of the image being scanned?

    > Your low quality slide at high resolution consists
    > of mostly grain noise and the image itself is very
    > soft. He hasn't lost much by downsampling it -
    > mostly edge detail around the dust.


    In the comparisons I posted, look at the detail on
    the cap, scarf, and on the texture of the coat. The
    loss is there, although not _very_ prominent.

    > The high frequency grain noise could be separated
    > from the true image with a bit of cunning adaptive
    > filtering.


    The addition of noise is an irreversible process,
    and all noice-removal tools (for both sound and
    images) operate probablity and do not have exact
    criteria of what it noise and what is not. They may
    partially help, but always at the expense of some
    real information. It is a fundamental law.

    Anton
     
    Anton Shepelev, Jul 6, 2011
    #16
    1. Advertising

Want to reply to this thread or ask your own question?

It takes just 2 minutes to sign up (and it's free!). Just click the sign up button to choose a username and then you can ask your own questions on the forum.
Similar Threads
  1. Silverstrand

    HEXUS.snapshots :: Xingtone ringtone creator

    Silverstrand, Sep 26, 2005, in forum: Front Page News
    Replies:
    0
    Views:
    580
    Silverstrand
    Sep 26, 2005
  2. scorpius

    Thunderbird revealing internal IPs

    scorpius, Nov 14, 2004, in forum: Firefox
    Replies:
    30
    Views:
    1,464
  3. jaygreg

    Revealing IP Addresses

    jaygreg, Aug 6, 2006, in forum: Computer Security
    Replies:
    4
    Views:
    597
    Sebastian Gottschalk
    Aug 6, 2006
  4. Radium

    I would like "" to answer my question

    Radium, Aug 25, 2007, in forum: Computer Support
    Replies:
    3
    Views:
    585
    =?ISO-8859-1?Q?R=F4g=EAr?=
    Aug 26, 2007
  5. RichA
    Replies:
    32
    Views:
    1,067
    Chris Malcolm
    Aug 18, 2009
Loading...

Share This Page