Re: [APML] Is DDP Necessary?

[<pleiades2004@pleiades-astrophoto.com>]

Hi Charles,

Perhaps you might want to read the original DDP information on Dr. Okano's
website:

http://www.asahi-net.or.jp/~rt6k-okn/viewgrap.htm

DDP was originally intended to achieve film-like appearance on CCD images.
CCD images are linear. For a given pixel in a CCD image, the relation
between brightness and exposure time is linear. This doesn't happen in film
images, which are wildly nonlinear.

The basic DDP procedure consists on a hyperbolic transfer function intended
to de-linearize linear (CCD) images, with the purpose of achieving a
response curve similar to the S-shaped gamma curve characteristic of film
images. By comparing the gamma curve to a linear function, you can see that
DDP can help in revealing image data at both ends of the dynamic range
(shadows and highlights).

DDP is just a shortcut to apply a relatively complex transfer curve. DDP can
of course be implemented as a curves adjustment. However, the advantage of
DDP is that it characterizes the underlying transfer curve by just a couple
of numeric parameters ('a' and 'b' in the original Okano's paper), which
simplifies things.

DDP can be applied with an embedded high-pass filtering effect. This is
called "edge emphasis" in Okano's paper.

Finally, DDP can also be applied to color images with a sort of "trick" to
improve color saturation with limited chrominance noise amplification. This
is called "color emphasis" in the original DDP terminology.

IMO, the bottom line is:

- If you work with film images, then actually DDP cannot help you, since
film is already nonlinear and there is no need to increase its nonlinearity.
Curves is all you need.

- If you work with CCD images, DDP can be helpful because it integrates some
processing steps into an easy-to-use black box. However, everything DDP does
can be done with curves, unsharp mask, and a well done color saturation
tool. IMHO, better results can be achieved when one controls each step
separately with specific processing tools.

There are many image processing software applications including DDP in their
arsenals. Just to mention some well-known examples: MaximDL, ImagesPlus,
StellaImage, AstroArt. Definitely you cannot use Photoshop because it lacks
division between images.

Regards,

Juan
__________________________________________________________________

Juan Conejero, Pleiades Astrophoto Team
juan.conejero__at__pleiades-astrophoto.com
PixInsight Home Page: http://pleiades-astrophoto.com/pixinsight/
__________________________________________________________________

----- Original Message ----- 
From: "Charles Shahar" <charles19@sympatico.ca>
To: <astro-photo@seds.org>
Sent: Tuesday, October 12, 2004 4:48 AM
Subject: [APML] Is DDP Necessary?


Hi Everyone:

I was wondering whether members feel Digital Development Process (DDP) is
necessary. I have become rather adept at pulling out faint details of
nebulosity using the PhotoShop curves and levels tools. Will a DDP filter be
more efficient in stretching out faint details, or should I trust my manual
tweaking more? If someone feels DDP will get better results, please let me
know what software you recommend for achieving it. Thanks,

-Charles


_______________________________________________
Astro-Photo mailing list
Astro-Photo@seds.org
http://seds.org/mailman/listinfo/astro-photo