High resolution...through digital interpolation...

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

My camera takes photos in normal resolution but claims to be able to take
finer photos at a higher pixel-rate through "digital interpolation".
Surely that's just stretching the image and not worth doing?
Can't I improve the image to the same degree later using filters in Corel
Photopaint?

Is there any real advantage in terms of image quality between an image
that's been digitally interpolated to a higher resolution?
It's no substitute for a higher resolution CCD in the camera is it?

D.
133 answers Last reply
More about high resolution through digital interpolation
  1. Archived from groups: rec.photo.digital (More info?)

    In article <d2ugk2$psk$1@sparta.btinternet.com>,
    plinioREMOVEdesignori@btopenworld.com says...
    > My camera takes photos in normal resolution but claims to be able to take
    > finer photos at a higher pixel-rate through "digital interpolation".
    > Surely that's just stretching the image and not worth doing?
    > Can't I improve the image to the same degree later using filters in Corel
    > Photopaint?
    >
    > Is there any real advantage in terms of image quality between an image
    > that's been digitally interpolated to a higher resolution?
    > It's no substitute for a higher resolution CCD in the camera is it?
    >
    > D.
    >
    >
    >

    On some cameras the improvement is there but only slight, on others its just
    an up-sizing that gives you a bigger picture, but not a better one.

    Fuji makes several that "interpolate upward" and on the S7000 there is
    USUALLY some improvement, but the file is twice as big, but NOT 2 times
    better. (at any rate, doubling the pixels dosent double the quality anyway,
    even if you increase the sensor count, twice as many isnt twice as good, only
    twice as big).


    --
    Larry Lynch
    Mystic, Ct.
  2. Archived from groups: rec.photo.digital (More info?)

    Des wrote:
    > My camera takes photos in normal resolution but claims to be able to take
    > finer photos at a higher pixel-rate through "digital interpolation".
    > Surely that's just stretching the image and not worth doing?
    > Can't I improve the image to the same degree later using filters in Corel
    > Photopaint?
    >
    > Is there any real advantage in terms of image quality between an image
    > that's been digitally interpolated to a higher resolution?
    > It's no substitute for a higher resolution CCD in the camera is it?
    >
    > D.
    >
    >
    It is marketing, just like 'digital zoom'. Many 'enhanced digital
    zooms' do basically the same thing, interpolating to a larger size, then
    cropping.
    It's mostly smoke and mirrors, but the pictures usually DO look
    smoother, just not more detailed.


    --
    Ron Hunter rphunter@charter.net
  3. Archived from groups: rec.photo.digital (More info?)

    Des wrote:
    > My camera takes photos in normal resolution but claims to be able to take
    > finer photos at a higher pixel-rate through "digital interpolation".
    > Surely that's just stretching the image and not worth doing?
    > Can't I improve the image to the same degree later using filters in Corel
    > Photopaint?
    >
    > Is there any real advantage in terms of image quality between an image
    > that's been digitally interpolated to a higher resolution?
    > It's no substitute for a higher resolution CCD in the camera is it?
    >
    > D.

    You have asked what you might call a loaded question, if you haven't
    noticed that already. From a technical point of view resolution is the
    smallest object that your camera can detect. This is determined by the
    number of pixels in the CCD of your camera (and the camera's optics).
    Interpolation CANNOT improve resolution. The only way to improve
    resolution is to have a CCD with more pixels. Some people will refer to
    resolution in respects to how good a picture looks. Although this isn't
    the technical definition of resolution it does get at the desired output
    of high resolution (i.e. nicer pictures). This may or maynot be
    improved with interpolation, but this kind of "pseudo-resolution" is in
    the eyes of the beholder.

    What interpolation does is increase the size of the image beyond what
    the CCD generates. The process which is used basically stretches out
    the image and makes educated guesses as to the color of the new pixels
    it creates. Although this creates a larger image, there is no more
    detail in the larger image then in the original image. In some cases
    interpolation may even decrease the overall results of the image due to
    the "guesses" it makes when it expands the image. Not only that, but
    interpolated images are larger, so you'll be able to take fewer pictures
    if you interpolate.

    As for what you should do in terms of taking pictures, my advice would
    be to take the pictures at your cameras maximum resolution without
    interpolation. This will let you get 100% of your camera's resolution,
    without wasting any space on your card. If you want to interpolate or
    otherwise alter your images afterwards you can do so with a graphics
    program such as Paint Shop Pro or Photoshop. These graphics programs
    have multiple other filters you can use, aside form interpolation, so
    you'll be able to adjust your images to a greater extent on-computer
    then you can on the camera.

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

    On Tue, 05 Apr 2005 17:07:15 +0000, Des wrote:

    > My camera takes photos in normal resolution but claims to be able to take
    > finer photos at a higher pixel-rate through "digital interpolation".
    > Surely that's just stretching the image and not worth doing?
    > Can't I improve the image to the same degree later using filters in Corel
    > Photopaint?
    >
    > Is there any real advantage in terms of image quality between an image
    > that's been digitally interpolated to a higher resolution?
    > It's no substitute for a higher resolution CCD in the camera is it?
    >
    > D.

    There's a simple rule on processing information
    (photo/video/audi) digitally:

    garbage in = garbage out

    It's as simple as that. Any pixel that was not produced by light hitting
    the sensor and having its colour properly registered, is not a real pixel.
    Sure it may be there, but it doesn't represent any visual information that
    was present in the scene when the photo was taken. It was calculated after
    the fact and therefore it is artificial.

    There is absolutely nothing that a digital process can do after the
    picture was taken to add real pixels. It is simply impossible. Now of
    course there are plenty of sharpening tricks that create a pretty good
    illusion of detail on interpolated images. They're nothing more than
    illusions though and you can only go a certain length before the fact that
    your pixels are fake starts showing. If the camera gives you all the
    pixels it actually captured, then it is in no way capable of more than
    good interpolating on your PC.. like Photopaint.

    In practice you can usually interpolate up to 125% or even 150% of the
    original size, apply a sharpening filter and end up with a decent print.
    The result however is no substitute for a higher real resolution.

    I see this a lot on really cheap desktop scanners. They claim to go up to
    14400dpi while their optics start to struggle beyond 600dpi. Years ago I
    bought a $3000,- flatbed scanner that had lesser specs than my neighbor's
    $100 scanner, at least when we compared the boxes they came in. Turned out
    that 'my' 1000dpi. was a lot sharper than 'his'. All the result of
    interpolation and good optics.

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

    Des writes:

    > My camera takes photos in normal resolution but claims to be able to take
    > finer photos at a higher pixel-rate through "digital interpolation".
    > Surely that's just stretching the image and not worth doing?

    Correct.

    > Can't I improve the image to the same degree later using filters in Corel
    > Photopaint?

    You can't improve an image through any type of manipulation. You will
    never have better quality than the image had when originally recorded.
    Interpolation is the creation of an optical illusion; it does not
    improve real image quality.

    > Is there any real advantage in terms of image quality between an image
    > that's been digitally interpolated to a higher resolution?

    No.

    > It's no substitute for a higher resolution CCD in the camera is it?

    No, it's not.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  6. Archived from groups: rec.photo.digital (More info?)

    Mxsmanic wrote:
    > Des writes:
    >
    >
    >>My camera takes photos in normal resolution but claims to be able to take
    >>finer photos at a higher pixel-rate through "digital interpolation".
    >>Surely that's just stretching the image and not worth doing?
    >
    >
    > Correct.
    >
    >
    >>Can't I improve the image to the same degree later using filters in Corel
    >>Photopaint?
    >
    >
    > You can't improve an image through any type of manipulation. You will
    > never have better quality than the image had when originally recorded.
    > Interpolation is the creation of an optical illusion; it does not
    > improve real image quality.
    >
    >
    >>Is there any real advantage in terms of image quality between an image
    >>that's been digitally interpolated to a higher resolution?
    >
    >
    > No.
    >
    >
    >>It's no substitute for a higher resolution CCD in the camera is it?
    >
    >
    > No, it's not.
    >

    Sigh.
    Ok, let's start a firestorm here.
    IF you have taken a picture at, say 4mp, and the picture has a lot of
    sharp lines, some at angles to the horizon, then you CAN get a better
    looking picture if you interpolate to a larger size, but ONLY because
    the interpolation algorithm is able to insert pictures that are the same
    as what would have been captured by a higher resolution sensor. They
    aren't 'real', but they end up in the same place as a real one would be,
    so the difference is rather more theoretical than practical.

    That said, the utility of this kind of interpolation is limited, and
    will rarely give you noticeably better results than just processing the
    picture with Photoshop.


    --
    Ron Hunter rphunter@charter.net
  7. Archived from groups: rec.photo.digital (More info?)

    Larry writes:

    > On some cameras the improvement is there but only slight, on others its just
    > an up-sizing that gives you a bigger picture, but not a better one.

    There is never an improvement.

    > Fuji makes several that "interpolate upward" and on the S7000 there is
    > USUALLY some improvement ...

    There is never any improvement. It's a mathematical impossibility.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  8. Archived from groups: rec.photo.digital (More info?)

    Mxsmanic wrote:
    > Larry writes:
    >
    >
    >>On some cameras the improvement is there but only slight, on others its just
    >>an up-sizing that gives you a bigger picture, but not a better one.
    >
    >
    > There is never an improvement.
    >
    >
    >>Fuji makes several that "interpolate upward" and on the S7000 there is
    >>USUALLY some improvement ...
    >
    >
    > There is never any improvement. It's a mathematical impossibility.
    >
    There is an 'apparent' improvement, since 'jaggies' are artificially
    reduced.

    --
    Ron Hunter rphunter@charter.net
  9. Archived from groups: rec.photo.digital (More info?)

    It is advertising bull. There is no such thing.

    --
    http://www.chapelhillnoir.com
    home of The Camera-ist's Manifesto
    The Improved Links Pages are at
    http://www.chapelhillnoir.com/links/mlinks00.html
    A sample chapter from "Haight-Ashbury" is at
    http://www.chapelhillnoir.com/writ/hait/hatitl.html

    "Des" <plinioREMOVEdesignori@btopenworld.com> wrote in message
    news:d2ugk2$psk$1@sparta.btinternet.com...
    > My camera takes photos in normal resolution but claims to be able to take
    > finer photos at a higher pixel-rate through "digital interpolation".
    > Surely that's just stretching the image and not worth doing?
    > Can't I improve the image to the same degree later using filters in Corel
    > Photopaint?
    >
    > Is there any real advantage in terms of image quality between an image
    > that's been digitally interpolated to a higher resolution?
    > It's no substitute for a higher resolution CCD in the camera is it?
    >
    > D.
    >
    >
  10. Archived from groups: rec.photo.digital (More info?)

    In message <f5n551dc9mak68tph6t1uqmm9ramecprpv@4ax.com>,
    Mxsmanic <mxsmanic@hotmail.com> wrote:

    >> Fuji makes several that "interpolate upward" and on the S7000 there is
    >> USUALLY some improvement ...

    >There is never any improvement. It's a mathematical impossibility.

    No, you must double the number of pixels in the output to preserve all
    the input; that is, if you intend to store in a traditional bitmap
    format.

    o o o o
    o o o
    o o o o
    o o o

    must become

    ooooooo
    ooooooo
    ooooooo
    ooooooo

    or detail will be lost.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <JPS@no.komm>
    ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
  11. Archived from groups: rec.photo.digital (More info?)

    In message <d2uuq1$894$1@news.ucalgary.ca>,
    Bryan Heit <bjheit@nospamucalgary.ca> wrote:

    >Interpolation CANNOT improve resolution.

    Yes, but the Fuji 7000, when it outputs 6MP from its 6MP sensor *is*
    interpolating, with loss. At 12MP, it is also interpolating, but with
    *no* loss.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <JPS@no.komm>
    ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
  12. Archived from groups: rec.photo.digital (More info?)

    Mxsmanic <mxsmanic@hotmail.com> wrote:

    >
    > You can't improve an image through any type of manipulation. You will
    > never have better quality than the image had when originally recorded.
    > Interpolation is the creation of an optical illusion; it does not
    > improve real image quality.
    >

    I suggest that you look into the field of deconvolution algorithms. Images
    CAN be improved. NASA/HST do it every day. Not to mention the NRO and a few
    other black ops.
  13. Archived from groups: rec.photo.digital (More info?)

    Bubbabob wrote:
    > Mxsmanic <mxsmanic@hotmail.com> wrote:
    >
    >
    >>You can't improve an image through any type of manipulation. You will
    >>never have better quality than the image had when originally recorded.
    >>Interpolation is the creation of an optical illusion; it does not
    >>improve real image quality.
    >>
    >
    >
    > I suggest that you look into the field of deconvolution algorithms. Images
    > CAN be improved. NASA/HST do it every day. Not to mention the NRO and a few
    > other black ops.

    yes, and if you had their processing power in your camera, you could
    probably do wonders too. Maybe NEXT year...


    --
    Ron Hunter rphunter@charter.net
  14. Archived from groups: rec.photo.digital (More info?)

    >I suggest that you look into the field of deconvolution algorithms. Images
    >CAN be improved.

    Images can be improved - we all do it when we lighten/darken etc but
    that is not interpolation.
    Interpolation (as I understand it) turns one pixel into four (for
    example) by "Guessing" what the colours should be.
    When nasa improve the picture the info is already there and they bring
    it to the fore.
  15. Archived from groups: rec.photo.digital (More info?)

    London@am2ma.eu wrote:
    >>I suggest that you look into the field of deconvolution algorithms. Images
    >>CAN be improved.
    >
    >
    > Images can be improved - we all do it when we lighten/darken etc but
    > that is not interpolation.
    > Interpolation (as I understand it) turns one pixel into four (for
    > example) by "Guessing" what the colours should be.
    > When nasa improve the picture the info is already there and they bring
    > it to the fore.
    >
    No, not always. Sometimes they combine information from several frames
    of the same data, taken with different filters, or different lighting,
    and the software makes educated guesses to fill in the blanks. It isn't
    necessarily EXACTLY what a better sensor would record, but it is
    PROBABLY what you would see if you were there.


    --
    Ron Hunter rphunter@charter.net
  16. Archived from groups: rec.photo.digital (More info?)

    JPS@no.komm wrote in news:sl3651p1vbv6i6clo8fvumcunbd4s56sus@4ax.com:

    > In message <d2uuq1$894$1@news.ucalgary.ca>,
    > Bryan Heit <bjheit@nospamucalgary.ca> wrote:
    >
    >>Interpolation CANNOT improve resolution.
    >
    > Yes, but the Fuji 7000, when it outputs 6MP from its 6MP sensor *is*
    > interpolating, with loss. At 12MP, it is also interpolating, but with
    > *no* loss.

    I'm not sure how you would arrive at that conclusion.

    It's much bigger brother the S3, for example, produces very good 6mp
    images, but it's 12 mp images aren't on par with those from a true 8MP
    body:

    http://www.dpreview.com/reviews/fujifilms3pro/page21.asp
  17. Archived from groups: rec.photo.digital (More info?)

    In message <Xns962FD3AD7A418ericvgillyahoocom@63.223.5.251>,
    Eric Gill <ericvgill@yahoo.com> wrote:

    >JPS@no.komm wrote in news:sl3651p1vbv6i6clo8fvumcunbd4s56sus@4ax.com:
    >
    >> In message <d2uuq1$894$1@news.ucalgary.ca>,
    >> Bryan Heit <bjheit@nospamucalgary.ca> wrote:
    >>
    >>>Interpolation CANNOT improve resolution.
    >>
    >> Yes, but the Fuji 7000, when it outputs 6MP from its 6MP sensor *is*
    >> interpolating, with loss. At 12MP, it is also interpolating, but with
    >> *no* loss.
    >
    >I'm not sure how you would arrive at that conclusion.

    Define "loss". Hint: it's not "quality of pixel units". It has more to
    do with relative absolutes between states.

    Using nearest neighbor (for example) to make a 9MP image out of a 1MP
    image is not "loss".

    >It's much bigger brother the S3, for example, produces very good 6mp
    >images, but it's 12 mp images aren't on par with those from a true 8MP
    >body:

    Probably for most things, but not for resolving horizontal and vertical
    edges.


    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <JPS@no.komm>
    ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
  18. Archived from groups: rec.photo.digital (More info?)

    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.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  19. Archived from groups: rec.photo.digital (More info?)

    Mxsmanic 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.
    >
    I have explained how it IS possible, and how it IS done. You would
    ignore a freight train as it mashed you to a pulp.


    --
    Ron Hunter rphunter@charter.net
  20. Archived from groups: rec.photo.digital (More info?)

    On Wed, 06 Apr 2005 06:58:36 +0200
    In message <33r6511j3p48qth3penttsq5ctslfn9o0h@4ax.com>
    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.

    Point of order: He said...

    Images CAN be improved.

    ....which is true.

    How much improvement is possible? That is the question.

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

    In message <33r6511j3p48qth3penttsq5ctslfn9o0h@4ax.com>,
    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.

    Define "it".

    You're talking about increasing maximum sampling resolution; they're
    talking about restoring contrast lost to MTF and AA filters.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <JPS@no.komm>
    ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
  22. Archived from groups: rec.photo.digital (More info?)

    London@am2ma.eu writes:

    > When nasa improve the picture the info is already there and they bring
    > it to the fore.

    Yes.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  23. Archived from groups: rec.photo.digital (More info?)

    Ron Hunter writes:

    > No, not always. Sometimes they combine information from several frames
    > of the same data, taken with different filters, or different lighting,
    > and the software makes educated guesses to fill in the blanks. It isn't
    > necessarily EXACTLY what a better sensor would record, but it is
    > PROBABLY what you would see if you were there.

    But in this case, also, they are using information obtained from the
    original capture. If you can capture more information at the source,
    you can improve an image.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  24. Archived from groups: rec.photo.digital (More info?)

    Ron Hunter wrote:
    > Bubbabob wrote:
    >
    >> Mxsmanic <mxsmanic@hotmail.com> wrote:
    >>
    >>
    >>> You can't improve an image through any type of manipulation. You will
    >>> never have better quality than the image had when originally recorded.
    >>> Interpolation is the creation of an optical illusion; it does not
    >>> improve real image quality.
    >>>
    >>
    >>
    >> I suggest that you look into the field of deconvolution algorithms.
    >> Images CAN be improved. NASA/HST do it every day. Not to mention the
    >> NRO and a few other black ops.
    >
    >
    > yes, and if you had their processing power in your camera, you could
    > probably do wonders too. Maybe NEXT year...
    >
    >
    Also, NASA and HST resolution improvement often hinges on having a good
    idea of what an object is, and optimizing algorithms that assume the
    shape of an object.

    This allows a powerful sharpening. BUT- one needs to be careful. If you
    use too much gain in these types of processes, even noise begins to look
    like the subject you assumed it was.

    An example. Lets say you are trying to improve a picture of a human
    face. You can determine what the spectral content (spatial spectrum,
    not temperal), and optimize algorithms for that spectral content.
    However, if you are not careful, this process can turn almost anything
    into something that looks like a human face.
  25. Archived from groups: rec.photo.digital (More info?)

    Mxsmanic wrote:

    > Bubbabob writes:
    >
    >>I suggest that you look into the field of deconvolution algorithms. Images
    >>CAN be improved.
    >
    > No, they cannot.

    Yes. They can. Typically on very high quality signal to noise data it is
    possible to obtain about a factor of 3x increase in apparent resolution
    on the brightest points using one of the regularised deconvolution
    methods like Maximum Entropy. The critical requirement is that you must
    know or be able to determine the blurring characteristics of the imaging
    system exactly in order to use them. It helps if everything is in the
    same focal plane - as is the case in astronomy where the techniques were
    developed.

    The question is posed in the form "what positive sky brightness
    distribution when measured by this equipment would give data consistent
    with the observations to within the measurement noise". It is routine in
    radio astronomy and frequently used in optical when the instrumental
    resolution is a limiting factor on the science.

    It has been possible since about 1978 and is more or less routine now in
    many fields of scientific endeavour. It is even within the capabilities
    of most home PCs and packages are available for amateur astronomers...

    >>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.

    No it isn't. Knowing a priori that image brightness is always positive
    is a tremendously powerful constraint on deconvolution algorithms.

    There will be some artefacts in any deconvolved image but there is also
    a better representation of what the target looks like as opposed to the
    conventional image as recorded by the sensor. They worry a great deal
    about validating these methods and cross checking - one such is:

    http://www.stsci.edu/stsci/meetings/irw/proceedings/briggsd.dir/briggsd.html

    Google: +deconvolution +superresolution +regularized
    will get you more especially in ADS abstracts.

    Regards,
    Martin Brown
  26. Archived from groups: rec.photo.digital (More info?)

    "Ron Hunter" <rphunter@charter.net> wrote in message
    news:x7C4e.1390$iW5.771@fe04.lga...
    > Sigh.
    > Ok, let's start a firestorm here.
    > IF you have taken a picture at, say 4mp, and the picture has a lot of
    > sharp lines, some at angles to the horizon, then you CAN get a better
    > looking picture if you interpolate to a larger size, but ONLY because the
    > interpolation algorithm is able to insert pictures that are the same as
    > what would have been captured by a higher resolution sensor. They aren't
    > 'real', but they end up in the same place as a real one would be, so the
    > difference is rather more theoretical than practical.
    >
    > That said, the utility of this kind of interpolation is limited, and will
    > rarely give you noticeably better results than just processing the picture
    > with Photoshop.
    >
    >
    > --
    > Ron Hunter rphunter@charter.net

    All are you always so far off the mark Ron?
    Just how do you think labs like www.fstoponline.com.au actually make 4 foot
    wide photographs from 6 Mp files? Maybe they do it with mirrors? No? Then
    maybe they do it with such poor quality, the clients of the few hundred
    Professional Photographers who use their service are all blind? No... I
    don't think that's the answer either. I know... They're Australians. Aussies
    don't know soap from clay do they? Geez, you can sell them bloody Aussies
    any line, eh? No you can't.

    They, like hundreds of other labs around the world use software to
    interpolate files as small as 3.5 MP and print them at 20"x30" every day
    with such stunning detail as they started out with, not even you could tell
    the prints were not just everyday photos from Medium Format film. Such is
    the world of digital enlargements as development passes you by.

    Don't go there Ron, you have no idea what your are talking about and until
    you do, all you will manage is to make yourself look stupid.
    http://users.tpg.com.au/hpc/examples2.htm

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

    Confused wrote:
    > On Wed, 06 Apr 2005 06:58:36 +0200
    > In message <33r6511j3p48qth3penttsq5ctslfn9o0h@4ax.com>
    > 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.
    >
    >
    > Point of order: He said...
    >
    > Images CAN be improved.
    >
    > ...which is true.
    >
    > How much improvement is possible? That is the question.
    >
    > Jeff
    How much improvement depends mainly on the subject matter. If it is
    mostly random shapes, such as trees, and grass, not a lot. Regular
    shapes with sharp lines and edges, quite a bit.
    The closer the interpolated pixels are to the reality of the scene, the
    more apparent improvement there will be.


    --
    Ron Hunter rphunter@charter.net
  28. Archived from groups: rec.photo.digital (More info?)

    Douglas wrote:
    > "Ron Hunter" <rphunter@charter.net> wrote in message
    > news:x7C4e.1390$iW5.771@fe04.lga...
    >
    >>Sigh.
    >>Ok, let's start a firestorm here.
    >>IF you have taken a picture at, say 4mp, and the picture has a lot of
    >>sharp lines, some at angles to the horizon, then you CAN get a better
    >>looking picture if you interpolate to a larger size, but ONLY because the
    >>interpolation algorithm is able to insert pictures that are the same as
    >>what would have been captured by a higher resolution sensor. They aren't
    >>'real', but they end up in the same place as a real one would be, so the
    >>difference is rather more theoretical than practical.
    >>
    >>That said, the utility of this kind of interpolation is limited, and will
    >>rarely give you noticeably better results than just processing the picture
    >>with Photoshop.
    >>
    >>
    >>--
    >>Ron Hunter rphunter@charter.net
    >
    >
    > All are you always so far off the mark Ron?
    > Just how do you think labs like www.fstoponline.com.au actually make 4 foot
    > wide photographs from 6 Mp files? Maybe they do it with mirrors? No? Then
    > maybe they do it with such poor quality, the clients of the few hundred
    > Professional Photographers who use their service are all blind? No... I
    > don't think that's the answer either. I know... They're Australians. Aussies
    > don't know soap from clay do they? Geez, you can sell them bloody Aussies
    > any line, eh? No you can't.
    >
    > They, like hundreds of other labs around the world use software to
    > interpolate files as small as 3.5 MP and print them at 20"x30" every day
    > with such stunning detail as they started out with, not even you could tell
    > the prints were not just everyday photos from Medium Format film. Such is
    > the world of digital enlargements as development passes you by.
    >
    > Don't go there Ron, you have no idea what your are talking about and until
    > you do, all you will manage is to make yourself look stupid.
    > http://users.tpg.com.au/hpc/examples2.htm
    >
    > Douglas
    >
    >
    I don't understand what you are arguing about. You said the same thing
    I did, then you disagree with me? Maybe it is you who doesn't know what
    he is talking about.

    So who looks stupid now?


    --
    Ron Hunter rphunter@charter.net
  29. Archived from groups: rec.photo.digital (More info?)

    Ron Hunter wrote:
    >
    > Douglas wrote:

    > > They, like hundreds of other labs around the world use software to
    > > print them at 20"x30" every day

    > > with such stunning detail as they started out with,

    and there is the true answer.

    what they started out with,
    and nothing more.

    to maintain visual sharpness,
    the ratio of megapixels must increase proportional to the area,
    or you must move back when you look at it.

    I have done 90x90 from 5 mp.
    looks great, and does what it was intended to do.

    <http://www.vircen.com/rpd/index.cgi?mode=image&album=/72dpi&image=Animal%20Connection018.JPG>

    but it was printed at 72dpi.


    > > Douglas
    > >
    > >
    > I don't understand what you are arguing about. You said the same thing
    > I did, then you disagree with me? Maybe it is you who doesn't know what
    > he is talking about.
    >
    > So who looks stupid now?
    >
    > --
    > Ron Hunter rphunter@charter.net
  30. Archived from groups: rec.photo.digital (More info?)

    Don Stauffer wrote:
    > Ron Hunter wrote:
    >
    >> Bubbabob wrote:
    >>
    >>> Mxsmanic <mxsmanic@hotmail.com> wrote:
    >>>
    >>>
    >>>> You can't improve an image through any type of manipulation. You will
    >>>> never have better quality than the image had when originally recorded.
    >>>> Interpolation is the creation of an optical illusion; it does not
    >>>> improve real image quality.
    >>>>
    >>>
    >>>
    >>> I suggest that you look into the field of deconvolution algorithms.
    >>> Images CAN be improved. NASA/HST do it every day. Not to mention the
    >>> NRO and a few other black ops.
    >>
    >>
    >>
    >> yes, and if you had their processing power in your camera, you could
    >> probably do wonders too. Maybe NEXT year...
    >>
    >>
    > Also, NASA and HST resolution improvement often hinges on having a good
    > idea of what an object is, and optimizing algorithms that assume the
    > shape of an object.
    >
    > This allows a powerful sharpening. BUT- one needs to be careful. If you
    > use too much gain in these types of processes, even noise begins to look
    > like the subject you assumed it was.
    >
    > An example. Lets say you are trying to improve a picture of a human
    > face. You can determine what the spectral content (spatial spectrum,
    > not temperal), and optimize algorithms for that spectral content.
    > However, if you are not careful, this process can turn almost anything
    > into something that looks like a human face.

    Like the human brain does? Grin.


    --
    Ron Hunter rphunter@charter.net
  31. Archived from groups: rec.photo.digital (More info?)

    "Eric Gill" <ericvgill@yahoo.com> wrote:
    > JPS@no.komm wrote in news:sl3651p1vbv6i6clo8fvumcunbd4s56sus@4ax.com:
    >
    > > In message <d2uuq1$894$1@news.ucalgary.ca>,
    > > Bryan Heit <bjheit@nospamucalgary.ca> wrote:
    > >
    > >>Interpolation CANNOT improve resolution.
    > >
    > > Yes, but the Fuji 7000, when it outputs 6MP from its 6MP sensor *is*
    > > interpolating, with loss. At 12MP, it is also interpolating, but with
    > > *no* loss.
    >
    > I'm not sure how you would arrive at that conclusion.

    The Fuji cameras have sensors that are rotated 45 degrees. What that means
    is that they can't be read out at their native resolution without losing
    information, and have to be read out at double resolution to retain all the
    information captured. The resulting image is a 12MP image that has 6MP of
    information.

    > It's much bigger brother the S3, for example, produces very good 6mp
    > images, but it's 12 mp images aren't on par with those from a true 8MP
    > body:
    >
    > http://www.dpreview.com/reviews/fujifilms3pro/page21.asp

    Exactly<g>. Fuji's idea that rotating the sensor increases resolution is, of
    course, hogwash.

    David J. Littleboy
    Tokyo, Japan
  32. Archived from groups: rec.photo.digital (More info?)

    David J. Littleboy wrote:


    > The Fuji cameras have sensors that are rotated 45 degrees.

    I don't see why this should automatically make a difference in the real world. I can
    see that having the pixel array parallel to lines in the subject woud make sense,
    such as a subject with lots of lines that are vertical and horizontal. OTOH, a
    picture of a pyramid would seem to be perfectly suited to a sensor that is rotated.

    I'll venture into areas on which I don't have all the details to touch on the OP's
    question and other responses.

    Anybody who has looked at a RAW image should be able to see that software
    interpolation can vastly improve on the unprocessed image that was actually captured.
    It's all (scientific) guesswork, but I think that any reasonable description of the
    result will be that it results in "finer detail" in the finished image even if some
    characterize it as artificial.

    If the algorithm can adjust the existig pixels based on adjacent pixels to create the
    processed image, there's no reason that new pixels can't be created with the same
    method. The only real debate should be about how many new pixels can be created
    before the unrealistic ones outnumber realistic ones enough to degrade the image. If
    a sensor captured an image of fine black lines on a white background, where each
    line was half the width of the pixel and the lines were perfectly parallel to the
    array I expect the result could be that half of the pixels would be black and half
    would be white. Or they could all be a uniform gray. To be interpolated (some people
    may need to check to see that the definition isn't "guessing", BTW) as a series of
    lines the algorithm would presumably have to have a built in bias to "look for" lines
    or edges and the pixels certainly couldn't be a uniform gray. How would the resulting
    image compare to the image of a field of dots that were all half the size of a pixel
    and perfectly centered on each pixel in the array? I realize that's completely
    unrealistic in the real world, but I'm guessing the resulting picture would be pretty
    unrealistic.

    The main point is that *all* processed images coming out of the camera are the result
    of interpolation, so it is clearly capable of improvement over the unprocessed values
    of individual pixels.

    --
    Steve

    The above can be construed as personal opinion in the absence of a reasonable
    belief that it was intended as a statement of fact.

    If you want a reply to reach me, remove the SPAMTRAP from the address.
  33. Archived from groups: rec.photo.digital (More info?)

    In article <d2ugk2$psk$1@sparta.btinternet.com>,
    "Des" <plinioREMOVEdesignori@btopenworld.com> wrote:

    > My camera takes photos in normal resolution but claims to be able to take
    > finer photos at a higher pixel-rate through "digital interpolation".
    > Surely that's just stretching the image and not worth doing?
    > Can't I improve the image to the same degree later using filters in Corel
    > Photopaint?
    >
    > Is there any real advantage in terms of image quality between an image
    > that's been digitally interpolated to a higher resolution?
    > It's no substitute for a higher resolution CCD in the camera is it?
    >
    Interpolation can be quite 'clever' or really dumb. It depends on how
    much time someone has spent on getting things right. However the fact
    remains - an 8mpxl camera will deliver a better image than a 6mpxl
    camera interpolated to 12mpxl. But that 12mpxl image can look as good as
    the 6mpxl and 12 megapixels will give more leeway when cropping and
    correcting perspective if you're trying to print A3. Filters and
    interpolation works in a similar way in PSP etc, but I feel the camera
    will do it better.

    The only way you can see if any advantage exists if a visual comparison
    - it's like the Dolby noise reduction for noisy audio cassettes - some
    people like it, other don't and if you, personally are satisfied with
    the quality, then it IS the better way.

    I'm sure the photo hacks working for print media will do 'anything' to
    get that shot.
  34. Archived from groups: rec.photo.digital (More info?)

    Ron Hunter wrote:
    >> An example. Lets say you are trying to improve a picture of a human
    >> face. You can determine what the spectral content (spatial spectrum,
    >> not temperal), and optimize algorithms for that spectral content.
    >> However, if you are not careful, this process can turn almost anything
    >> into something that looks like a human face.
    >
    >
    > Like the human brain does? Grin.
    >
    >
    Yep, I didn't want to say it, but since you did, I certainly agree. We
    are wired that way- the algorithm is built into the old Mk 1 brain :-)
  35. Archived from groups: rec.photo.digital (More info?)

    Ron Hunter <rphunter@charter.net> writes:

    >> The 6 MP Fuji cameras are 6 MP cameras that happen to require the use of
    >> a 12 MP output file to convert between diagonal and row/column image
    >> form without loss of information.

    >Rather like a scanner with 600dpi optical in the scan across the field,
    >and 1200dpi in the long side as the stepper motor makes 1/1200" steps?
    >Works for me.

    Well, vaguely similar. The scanner in 1200 DPI mode is capturing
    more *real* information in the vertical direction, although at low
    contrast. The Fuji cameras capture no more nor less information than
    other 6 MP cameras; they just maximize resolution in a different
    direction.

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

    "Des" <plinioREMOVEdesignori@btopenworld.com> wrote in message
    news:d2ugk2$psk$1@sparta.btinternet.com...
    > My camera takes photos in normal resolution but claims to be able to take
    > finer photos at a higher pixel-rate through "digital interpolation".
    > Surely that's just stretching the image and not worth doing?
    > Can't I improve the image to the same degree later using filters in Corel
    > Photopaint?
    >
    > Is there any real advantage in terms of image quality between an image
    > that's been digitally interpolated to a higher resolution?
    > It's no substitute for a higher resolution CCD in the camera is it?
    >
    > D.

    Thanks for all the postings on my initial enquiry; reading them has
    confirmed me in my intentions to use the camera's normal resolution and do
    the improvement later in Corel PhotoPaint.
    Of course I can't add information where none has been recorded during the
    initial exposure, but here's what I'd do once I had the image.
    I'd load the image to Corel, and apply Noise/Remove Moiré, followed by
    Remove Noise, then a couple of passes of 80% Adaptive Unsharp.
    Then I'd turn my attention to the contrast/colour/brightness and tweak them
    to increase the apparent clarity of the image. From there (if I wanted to go
    further) I'd seek out features in the image that were capable of being
    enhanced/straightened by applying straight-edged (or segments of the
    circular shaped) masks, sampling the colours and making the edges crisp and
    smooth.

    I accept that I'm not `bringing out' extra information from the camera's
    image; we've already established that's impossible.
    It's possible to add more detail using information from the user's brain,
    though, and to produce an image more representative of the subject than was
    originally available through the camera alone.

    Thank you folks!

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

    On Wed, 06 Apr 2005 10:09:33 -0500, Ron Hunter <rphunter@charter.net>
    wrote:

    >Confused wrote:
    >
    >> How much improvement is possible? That is the question.
    >>
    >How much improvement depends mainly on the subject matter. If it is
    >mostly random shapes, such as trees, and grass, not a lot.

    Disagree. These are excellent subjects for fractal representation.

    > Regular
    >shapes with sharp lines and edges, quite a bit.

    The vector approach works well here, yes.

    --
    Owamanga!
    http://www.pbase.com/owamanga
  38. Archived from groups: rec.photo.digital (More info?)

    Ron Hunter writes:

    > And if you took a picture of something that was all vertical and
    > horizontal lines?

    I'm not interested in test charts.

    > You could then interpolate to just about any level,
    > and the picture would be an accurate representation of the original, and
    > the same as from a camera with whatever resolution you could find.

    Really? Try doing that with a picture of a picket fence.>

    > This
    > is a rather limited case, of course, but it does illustrate the point.
    > IF the subject matter lends itself to interpolation, then much
    > improvement, indistinguishable from 'real' can be had.
    > So, what does your information theory have to say about that?

    That there is no net increase in information.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  39. Archived from groups: rec.photo.digital (More info?)

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

    > This is a very narrow-minded view of the problem.

    It's a harsh reality.

    > Your motto: "They said it couldn't be done, so I didn't try".

    It cannot be done, no matter how much one tries.

    > If objects can be correctly identified by software, they can be
    > re-rendered at any resolution.

    Only if the software contains all the detail concerning the objects. In
    other words, only if all the detail is already present (the fact that
    the software contains it instead of the captured image doesn't alter the
    constraints of information theory).

    In practice, no software can do this outside of the most trivial test
    cases.

    > Image recognition can read the text, identify each
    > letter, identify the font used and re-render it at 100 times the
    > original, maintaining the angle, color balance and texture from the
    > original.

    Some fonts are so slightly different that they cannot be identified in
    this way.

    > Software
    > development moves fast, and even though we may not have the magic
    > 'enlarge' button in Photoshop yet, it *will* be there one day.

    No, it won't.

    What we will have is capture at higher resolutions instead.

    This reminds me of a claim I heard from someone long ago who said that
    the future would be shaped by ever-improving compression algorithms. In
    fact, it has been shaped by ever-increasing bandwidth.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  40. Archived from groups: rec.photo.digital (More info?)

    On Wed, 06 Apr 2005 17:47:09 +0200, Mxsmanic <mxsmanic@hotmail.com>
    wrote:


    >> Your motto: "They said it couldn't be done, so I didn't try".
    >
    >It cannot be done, no matter how much one tries.

    I seem to recall this was said about powered flight, too.

    --
    Bill Funk
    Change "g" to "a"
  41. Archived from groups: rec.photo.digital (More info?)

    Don Stauffer writes:

    > Also, NASA and HST resolution improvement often hinges on having a good
    > idea of what an object is, and optimizing algorithms that assume the
    > shape of an object.

    But that's just another way of saying that they have additional image
    data.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  42. Archived from groups: rec.photo.digital (More info?)

    On Wed, 06 Apr 2005 17:48:21 +0200, Mxsmanic <mxsmanic@hotmail.com>
    wrote:

    >Don Stauffer writes:
    >
    >> Also, NASA and HST resolution improvement often hinges on having a good
    >> idea of what an object is, and optimizing algorithms that assume the
    >> shape of an object.
    >
    >But that's just another way of saying that they have additional image
    >data.

    It's another way of saying NASA have an open minded approach to
    problem solving.

    --
    Owamanga!
    http://www.pbase.com/owamanga
  43. Archived from groups: rec.photo.digital (More info?)

    Ron Hunter writes:

    > I have explained how it IS possible, and how it IS done.

    You've overlooked the covert channel, as I've pointed out in a separate
    post.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  44. Archived from groups: rec.photo.digital (More info?)

    JPS@no.komm writes:

    > Define "it".

    Producing an image that contains more detail than was available in the
    original capture.

    > You're talking about increasing maximum sampling resolution; they're
    > talking about restoring contrast lost to MTF and AA filters.

    You cannot restore contrast, you can only simulate it.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  45. Archived from groups: rec.photo.digital (More info?)

    Martin Brown writes:

    > Yes. They can. Typically on very high quality signal to noise data it is
    > possible to obtain about a factor of 3x increase in apparent resolution
    > on the brightest points using one of the regularised deconvolution
    > methods like Maximum Entropy.

    Apparent resolution is not actual detail.

    > The critical requirement is that you must
    > know or be able to determine the blurring characteristics of the imaging
    > system exactly in order to use them.

    If you know enough to fully reconstruct missing detail in the image, you
    don't need the image in the first place.

    > No it isn't. Knowing a priori that image brightness is always positive
    > is a tremendously powerful constraint on deconvolution algorithms.

    Knowing anything in advance adds image information.

    If that advance knowledge doesn't match the reality of the original
    scene, though, the results can be hugely misleading.

    --
    Transpose hotmail and mxsmanic in my e-mail address to reach me directly.
  46. Archived from groups: rec.photo.digital (More info?)

    JPS@no.komm wrote:
    > In message <d2uuq1$894$1@news.ucalgary.ca>,
    > Bryan Heit <bjheit@nospamucalgary.ca> wrote:
    >
    >
    >>Interpolation CANNOT improve resolution.
    >
    >
    > Yes, but the Fuji 7000, when it outputs 6MP from its 6MP sensor *is*
    > interpolating, with loss. At 12MP, it is also interpolating, but with
    > *no* loss.

    This seems to be a specific issue with fuji cameras, whom for some
    reason rotated their CCD which defies all logic when it comes to
    imaging. The large majority of camera will not benefit from
    interpolation because they are designed properly, and can read the full
    CCD resolution without interpolation. I cannot think of a single reason
    why fuji would do this, aside from some sort of rip-off marketing gimic.

    I do a lot of imaging for scientific purposes (predominantly
    microscopy), and we avoid any image interpolation due to it's tendency
    to induce artifacts. I take this knowledge into my photography as I
    know that the uninterpolated image I produce will be truer then an
    interpolated image. Afterwards I can alter the image as I see fit -
    interpolation, cropping, etc, but by not interpolating on the camera you
    give yourself the maximal ability to alter your image as what you start
    with is the raw, completely unprocessed image. Once interpolated it is
    impossible to get back to the raw image.

    Keep in mind that the raw image is the maximal amount of detail your
    camera can capture - interpolation can produce a larger image with the
    appearance of greater resolution, but in reality the image is no more
    detailed (and possible less detailed) then the original image.

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

    In message <q71851p1ohrtb9ehrq7q0c58go763lldrd@4ax.com>,
    Mxsmanic <mxsmanic@hotmail.com> wrote:

    >JPS@no.komm writes:

    >> You're talking about increasing maximum sampling resolution; they're
    >> talking about restoring contrast lost to MTF and AA filters.
    >
    >You cannot restore contrast, you can only simulate it.

    If you know how much the pixel-to-pixel contrast is attenuated, you can
    bring it back up again, albeit by increasing the capture artifacts as
    well.
    --

    <>>< ><<> ><<> <>>< ><<> <>>< <>>< ><<>
    John P Sheehy <JPS@no.komm>
    ><<> <>>< <>>< ><<> <>>< ><<> ><<> <>><
  48. Archived from groups: rec.photo.digital (More info?)

    Mxsmanic wrote:

    > Martin Brown writes:
    >
    >>Yes. They can. Typically on very high quality signal to noise data it is
    >>possible to obtain about a factor of 3x increase in apparent resolution
    >>on the brightest points using one of the regularised deconvolution
    >>methods like Maximum Entropy.
    >
    > Apparent resolution is not actual detail.

    Some of it is. You can know the positions and relative positions of
    bright point sources much more accurately than to the nearest pixel even
    with relatively crude deconvolution methods.

    >>The critical requirement is that you must
    >>know or be able to determine the blurring characteristics of the imaging
    >>system exactly in order to use them.
    >
    > If you know enough to fully reconstruct missing detail in the image, you
    > don't need the image in the first place.

    Rubbish. You only need to know the point spread function. And the
    positivity constraint - but there are still a lot of all positive images.
    >
    >>No it isn't. Knowing a priori that image brightness is always positive
    >>is a tremendously powerful constraint on deconvolution algorithms.
    >
    > Knowing anything in advance adds image information.

    No it adds additional information beyond the actual raw image. Namely:

    The point spread function that the original target image was convolved
    with to make the raw data.

    And usually that there can be no regions of negative intensity in the
    real world.

    Taken together these provide the basis for genuine superresolution.
    >
    > If that advance knowledge doesn't match the reality of the original
    > scene, though, the results can be hugely misleading.

    The *fundamental* point that you are missing (perhaps by being
    deliberately obtuse) is that there are never any negative brightness
    regions in the real world. We sense things by light arriving at the
    detector. This a priori knowledge provides the basis for most of the
    enhanced resolution achieved by modern deconvolution codes and it is
    very real from an information theoretic point of view.

    Regions of frequency space where no data was measured can be
    reconstructed reliably by imposing the positivity constraint. The answer
    may not be perfect but it is a heck of a lot better than the raw image.

    Regards,
    Martin Brown
  49. Archived from groups: rec.photo.digital (More info?)

    On Wed, 06 Apr 2005 09:47:42 GMT
    In message <2fO4e.1779$5F3.1443@news-server.bigpond.net.au>
    Posted from BigPond Internet Services
    "Douglas" <decipleofeos@hotmail.com> wrote:

    > ...
    > http://users.tpg.com.au/hpc/examples2.htm

    "...harnessing the power of several powerful microprocessors."

    Please, edumacate us. What programs and processors?

    Jeff
Ask a new question

Read More

Photo Cameras Resolution