We need standards

Discussion in 'Digital Photography' started by Mark Herring, Nov 4, 2003.

  1. Mark Herring

    Mark Herring Guest

    Digital friends;

    I would like to open a dialog on how we can move towards some
    standards for how digital imaging technology is specified. The
    regulars here will quickly see that I am motivated in large part by
    the rather amazing debates about the merits of the FoveonX3 and Fuji
    "Super-CCD" technology.

    The most urgent need, in my view, is to somehow capture what I have
    called the "de-facto" standard for how spatial resolution is
    specified---ie the number of true, spatially independent, samples that
    are acquired by the sensor. It is an imperfect definition, in that it
    only gives you the limiting resolution---ie the best that the device
    can be if all other factors are perfect. It does, however, allow some
    rational basis for comparing cameras.

    In addition to standards for specifications, would it not also be
    useful to have some standard tests---ones that could be duplicated
    anywhere? Such a test would require that something be photographed
    under controlled conditions. The "penny test" is a step in this
    direction, but random pennies pulled out of one's pocket are all
    different, and the controlled conditions are not specified.

    Does anyone know if any of the traditional organizations---eg ISO--are
    dealing with this issue?...or plan to?

    Any ideas for what the standards should be?

    -Mark
    **************************
    Mark Herring, Pasadena, Calif.
    Private e-mail: Just say no to "No".
    Mark Herring, Nov 4, 2003
    #1
    1. Advertising

  2. Mark Herring

    Tony Spadaro Guest

    What matters what anyone sets as a standard since any company that wishes
    to inflate sales will ignore those stnadards and continue to come to the
    conclusion that 3.4 equals 10.
    Thirty years ago there were a half dozen ways to measure the output of a
    HiFi amplifier. After a long battle the RMS system was declared to tbe the
    "standard" by which all such amplifiers are measuered. Today the boxes
    simply state an output in watts - with no explanation of what measurement
    was used. Companies simply ignored the RMS "standard" and put up any
    inflated number they pleased.

    --
    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 my novel "Haight-Ashbury" is at
    http://www.chapelhillnoir.com/writ/hait/hatitl.html
    "Mark Herring" <> wrote in message
    news:...
    > Digital friends;
    >
    > I would like to open a dialog on how we can move towards some
    > standards for how digital imaging technology is specified. The
    > regulars here will quickly see that I am motivated in large part by
    > the rather amazing debates about the merits of the FoveonX3 and Fuji
    > "Super-CCD" technology.
    >
    > The most urgent need, in my view, is to somehow capture what I have
    > called the "de-facto" standard for how spatial resolution is
    > specified---ie the number of true, spatially independent, samples that
    > are acquired by the sensor. It is an imperfect definition, in that it
    > only gives you the limiting resolution---ie the best that the device
    > can be if all other factors are perfect. It does, however, allow some
    > rational basis for comparing cameras.
    >
    > In addition to standards for specifications, would it not also be
    > useful to have some standard tests---ones that could be duplicated
    > anywhere? Such a test would require that something be photographed
    > under controlled conditions. The "penny test" is a step in this
    > direction, but random pennies pulled out of one's pocket are all
    > different, and the controlled conditions are not specified.
    >
    > Does anyone know if any of the traditional organizations---eg ISO--are
    > dealing with this issue?...or plan to?
    >
    > Any ideas for what the standards should be?
    >
    > -Mark
    > **************************
    > Mark Herring, Pasadena, Calif.
    > Private e-mail: Just say no to "No".
    >
    Tony Spadaro, Nov 4, 2003
    #2
    1. Advertising

  3. Mark Herring

    Samuel Paik Guest

    Mark Herring <> wrote:
    > The most urgent need, in my view, is to somehow capture what I have
    > called the "de-facto" standard for how spatial resolution is
    > specified---ie the number of true, spatially independent, samples that
    > are acquired by the sensor. It is an imperfect definition, in that it
    > only gives you the limiting resolution---ie the best that the device
    > can be if all other factors are perfect. It does, however, allow some
    > rational basis for comparing cameras.


    Standard? It's pretty easy to get these numbers for most digital
    image sensors. The oddball one is for the various Fuji "weirdos".

    > In addition to standards for specifications, would it not also be
    > useful to have some standard tests---ones that could be duplicated
    > anywhere?


    There are standard ISO test targets. Both Dpreview and Imaging
    Resource routinely take photos of the basic resolution chart
    in their reviews, although they don't set up completely precisely,
    they do a reasonable enough job to give readers a good idea.
    (BTW, the basic chart is inadequate--does not supply close enough
    lines--for the highest resolution DSLRs available today, e.g.
    EOS-1Ds, Kodak DCS-14n).

    Various organizations have even more sophisticated tests available.
    Samuel Paik, Nov 4, 2003
    #3
  4. Mark Herring

    Don Stauffer Guest

    There is a fine standard for just that, an ISO standard. Problem is, no
    one uses it. Mfgrs sure don't. And mags reporting on cameras do not
    do the test specified in the standard. I think the problem is, how do we
    get people to use the existing standard.

    Mark Herring wrote:
    >
    > Digital friends;
    >
    > I would like to open a dialog on how we can move towards some
    > standards for how digital imaging technology is specified. The
    > regulars here will quickly see that I am motivated in large part by
    > the rather amazing debates about the merits of the FoveonX3 and Fuji
    > "Super-CCD" technology.
    >
    > The most urgent need, in my view, is to somehow capture what I have
    > called the "de-facto" standard for how spatial resolution is
    > specified---ie the number of true, spatially independent, samples that
    > are acquired by the sensor. It is an imperfect definition, in that it
    > only gives you the limiting resolution---ie the best that the device
    > can be if all other factors are perfect. It does, however, allow some
    > rational basis for comparing cameras.
    >
    > In addition to standards for specifications, would it not also be
    > useful to have some standard tests---ones that could be duplicated
    > anywhere? Such a test would require that something be photographed
    > under controlled conditions. The "penny test" is a step in this
    > direction, but random pennies pulled out of one's pocket are all
    > different, and the controlled conditions are not specified.
    >
    > Does anyone know if any of the traditional organizations---eg ISO--are
    > dealing with this issue?...or plan to?
    >
    > Any ideas for what the standards should be?
    >
    > -Mark
    > **************************
    > Mark Herring, Pasadena, Calif.
    > Private e-mail: Just say no to "No".


    --
    Don Stauffer in Minnesota

    webpage- http://www.usfamily.net/web/stauffer
    Don Stauffer, Nov 4, 2003
    #4
  5. "Mark Herring" <> wrote in message
    news:...
    SNIP
    > Does anyone know if any of the traditional organizations---eg ISO--are
    > dealing with this issue?...or plan to?


    In addition to available standards, the ISO is finalizing several standards
    for "Electronic still-picture cameras". But of course, anyone can choose to
    ignore them.

    > Any ideas for what the standards should be?


    ISO 12233:2000 Photography -- Electronic still-picture cameras --
    Resolution measurements (available in English only):
    (http://www.iso.org/iso/en/CatalogueDetailPage.CatalogueDetail?CSNUMBER=3371
    5&ICS1=37&ICS2=40&ICS3=10).
    For CHF 114,00 it's yours (PDF or paper version).

    The ISO is very clear about pixels not being a measure of resolution, which
    makes sense because a pixel is dimensionless until output to a physical
    medium. When pixels are mentioned, they are often referred to as "luminance
    pixels".

    They do specify the following:
    - Resolution: a measure of the ability of a digital image capture system,
    or a component of a digital image capture system, to capture fine spatial
    detail. Resolution measurement metrics include resolving power, limiting
    visual resolution, SFR, MTF and CTF.
    - Sampled imaging system: an imaging system or device which generates an
    image signal by sampling an image at an array of discrete points, or along a
    set of discrete lines, rather than a continuum of points. The sampling at
    each point is done using a finite size sampling aperture or area.
    - Sample spacing: the physical distance between sampling points or sampling
    lines. The sample spacing may be different in the two orthogonal sampling
    directions. Measured in units of distance (e.g., microns, mm.)
    - Sampling frequency: the reciprocal of sample spacing. Expressed in samples
    per unit distance. (e.g. dots per inch, DPI)

    Standards also relevant for this subject are:
    ISO 12231:1997 Photography -- Electronic still-picture cameras --
    Terminology (available in English only). Note: A new version is in the works
    (ISO/DIS 12231-2)
    ISO 14524:1999 Photography -- Electronic still-picture cameras -- Methods
    for measuring opto-electronic conversion functions (OECFs) (available in
    English only)
    ISO 15739:2003 Photography -- Electronic still-picture imaging -- Noise
    measurements (available in English only)

    and for current projects:
    <http://www.iso.ch/iso/en/stdsdevelopment/techprog/workprog/TechnicalProgram
    meTCDetailPage.TechnicalProgrammeTCDetail?COMMID=1603>
    Careful, the link probably wraps to two lines.

    Bart
    Bart van der Wolf, Nov 4, 2003
    #5
  6. Mark Herring

    Mark Herring Guest

    On Tue, 04 Nov 2003 08:08:40 -0600, Don Stauffer
    <> wrote:

    >There is a fine standard for just that, an ISO standard. Problem is, no
    >one uses it. Mfgrs sure don't. And mags reporting on cameras do not
    >do the test specified in the standard. I think the problem is, how do we
    >get people to use the existing standard.
    >


    Don;

    What standard are you referring to?

    What I am looking for is something simple like ISO film speeds. ISO
    400 gives my a pretty good picture of what a film will do.

    Just as an example, If we had a standard for maximum resolution that
    was unambiguous, then it might not be that hard to enforce it--at
    least in reviews, newgroup discussions, etc.

    _Mark
    **************************
    Mark Herring, Pasadena, Calif.
    Private e-mail: Just say no to "No".
    Mark Herring, Nov 5, 2003
    #6
  7. Mark Herring

    Mark Herring Guest

    On Tue, 4 Nov 2003 15:38:08 +0100, "Bart van der Wolf"
    <> wrote:


    >ISO 12233:2000 Photography -- Electronic still-picture cameras --
    >Resolution measurements (available in English only):
    >(http://www.iso.org/iso/en/CatalogueDetailPage.CatalogueDetail?CSNUMBER=3371
    >5&ICS1=37&ICS2=40&ICS3=10).
    >For CHF 114,00 it's yours (PDF or paper version).
    >


    Lots snipped.......
    >
    >and for current projects:
    ><http://www.iso.ch/iso/en/stdsdevelopment/techprog/workprog/TechnicalProgram
    >meTCDetailPage.TechnicalProgrammeTCDetail?COMMID=1603>
    >Careful, the link probably wraps to two lines.
    >
    >Bart
    >

    Bart;

    Thanks for all the info---can't absorb it at the moment. In all this,
    do you see any simple standard for resolution---akin to ISO film
    speed?

    -Mark
    **************************
    Mark Herring, Pasadena, Calif.
    Private e-mail: Just say no to "No".
    Mark Herring, Nov 5, 2003
    #7
  8. Mark Herring

    Don Stauffer Guest

    Indeed, as defined in earlier image technologies, resolution is the
    quantified ability to do a specific task task, and is the result of
    tests. Normally in photographic and TV work, it is the ability to
    recognize that a series of alternating dark and light bars are just
    that, not a grey patch.

    Bart van der Wolf wrote:

    >
    > The ISO is very clear about pixels not being a measure of resolution, which
    > makes sense because a pixel is dimensionless until output to a physical
    > medium. When pixels are mentioned, they are often referred to as "luminance
    > pixels".
    >
    >
    > Bart


    --
    Don Stauffer in Minnesota

    webpage- http://www.usfamily.net/web/stauffer
    Don Stauffer, Nov 5, 2003
    #8
  9. Mark Herring

    Frank ess Guest

    "Mark Herring" <> wrote in message
    news:...
    > On Tue, 4 Nov 2003 15:38:08 +0100, "Bart van der Wolf"
    > <> wrote:
    >
    >
    > >ISO 12233:2000 Photography -- Electronic still-picture cameras --
    > >Resolution measurements (available in English only):

    >
    >(http://www.iso.org/iso/en/CatalogueDetailPage.CatalogueDetail?CSNUMBER=337

    1
    > >5&ICS1=37&ICS2=40&ICS3=10).
    > >For CHF 114,00 it's yours (PDF or paper version).
    > >

    >
    > Lots snipped.......
    > >
    > >and for current projects:

    >
    ><http://www.iso.ch/iso/en/stdsdevelopment/techprog/workprog/TechnicalProgra

    m
    > >meTCDetailPage.TechnicalProgrammeTCDetail?COMMID=1603>
    > >Careful, the link probably wraps to two lines.
    > >
    > >Bart
    > >

    > Bart;
    >
    > Thanks for all the info---can't absorb it at the moment. In all this,
    > do you see any simple standard for resolution---akin to ISO film
    > speed?
    >


    I think it is on the schedule just after "audio-Watts-of-power" is resolved.
    Frank ess, Nov 5, 2003
    #9
  10. Mark Herring

    Guest

    The nice thing about standards is that we have so many from which to
    choose.
    , Nov 5, 2003
    #10
  11. "Mark Herring" <> wrote in message
    news:...
    > On Tue, 4 Nov 2003 15:38:08 +0100, "Bart van der Wolf"
    > <> wrote:
    >
    >
    > >ISO 12233:2000 Photography -- Electronic still-picture cameras --
    > >Resolution measurements (available in English only):

    SNIP

    > Thanks for all the info---can't absorb it at the moment. In all this,
    > do you see any simple standard for resolution---akin to ISO film
    > speed?


    The ISO "slanted edge" method is pretty simple, it only needs some
    mathematical evaluation to get the results (MTF like graphs, called SFR or
    Spatial Frequency Response curves). It will take some programming to speed
    up the evaluation. I may be able to find a method using free software.

    An alternative can be home made with little effort. I use a relatively
    simple star target, that gives both a visual impression, and the possibility
    to calculate the limiting resolution or an MTF (requires to take multiple
    measurements). It is easy to set up, use, and the interpretation is also
    very simple.

    Bart
    Bart van der Wolf, Nov 5, 2003
    #11
  12. Re: We need standards - Resolution

    "Bart van der Wolf" <> wrote in message
    news:3fa98ed7$0$58702$4all.nl...
    SNIP
    > An alternative can be home made with little effort. I use a relatively
    > simple star target, that gives both a visual impression, and the

    possibility
    > to calculate the limiting resolution or an MTF (requires to take multiple
    > measurements). It is easy to set up, use, and the interpretation is also
    > very simple.


    I've just uploaded the file (2 versions) for download.
    For HP inkjet printers (3.77MB):
    http://www.xs4all.nl/~bvdwolf/main/downloads/Jtf60cy-100mm_600ppi.gif
    For Epson inkjet printers (5.28MB):
    http://www.xs4all.nl/~bvdwolf/main/downloads/Jtf60cy-100mm_720ppi.gif

    Print this file unaltered (switch-off printer enhancements) at the indicated
    ppi setting (use glossy paper for output). It should result in a print of
    100x100mm.

    Shoot an image of the target from a couple of meters distance (exact
    distance is not critical, 2-3 meters is fine). Using a tripod, mirror
    lock-up, and stopping down your lens to the optimal aperture will benefit
    the result. You may want to shoot several frames to eliminate autofocus
    variance and afterwards choose the best one (digital is cheap).

    Evaluation:
    1. First you need to determine the sampling pitch (center to center distance
    between photosites in a row or column) of your sensor.
    For example, if you scan film with a 4000ppi scanner, the pitch is 1/4000
    inch, or 25.4/4000=0.00635 mm per output pixel, if you use a Canon D60 /
    10D, then the pitch is about 0.007377 mm, a 1Ds has a pitch of 0.0088 mm, a
    Sigma SD9/SD10 has a pitch of 0.00912 mm, etc., etc..

    2. Open the image in your photoeditor, and zoom in (e.g. 300%) on the
    target. You can approximately color balance the target by setting the
    blackest point in your target image close to RGB 5,5,5 and the whitest point
    to RGB 240,240,240. Adjusting the midpoint (target background) to RGB
    127,127,127 should then give a neutral balance with approximately the right
    contrast for the entire image.

    3. The star target image consists of 60 full modulation cycles from black to
    white, on a mid gray background. The circumference of any concentric circle
    will also equal 60 cycles. Taking Pi (3.14159...) and the pitch in
    millimeters you determined in step 1, this means that for your sensor 1
    cycle/mm = 60/(Pi*pitch) pixels (you may approximate that with 19.1/pitch).

    In your photo editor you now measure the diameter of the unresolved center
    in pixels. Now divide the number of pixels you found for 1 cy/mm by the
    diameter in pixels you measured, and the result is the limiting resolution
    in cy/mm or lp/mm for your entire imaging chain.

    As long as you shoot with the same sensor pitch, you can use the same
    calculation and only divide by the newly measured diameter. The blur
    diameter on the sensor will not change with target distance, because the
    resolution is the limiting factor and not the target magnification.

    This will enable to find the differences between lenses, zoom settings, and
    apertures for *your* imaging chain. It will also help to estimate the amount
    of magnification you can apply to your images before the viewing resolution
    is compromised.

    Bart
    Bart van der Wolf, Nov 6, 2003
    #12
  13. David J. Littleboy, Nov 6, 2003
    #13
    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. Michael Ray Brown

    Picture CD standards

    Michael Ray Brown, Aug 24, 2003, in forum: Digital Photography
    Replies:
    5
    Views:
    546
    Michael Ray Brown
    Aug 27, 2003
  2. LesU

    costco standards

    LesU, Sep 8, 2003, in forum: Digital Photography
    Replies:
    12
    Views:
    1,181
    mark_digitalĀ©
    Sep 13, 2003
  3. schuetzen - RKBA!

    What minimum standards should exist for a film scanner??

    schuetzen - RKBA!, Jul 11, 2004, in forum: Digital Photography
    Replies:
    3
    Views:
    285
    M Barnes
    Jul 12, 2004
  4. John

    Star Wars & double standards

    John, Feb 11, 2004, in forum: DVD Video
    Replies:
    83
    Views:
    1,807
    Mike Kohary
    Feb 22, 2004
  5. Den

    Video Standards Converter

    Den, Apr 4, 2004, in forum: DVD Video
    Replies:
    2
    Views:
    506
Loading...

Share This Page