Howdy, folks,
I'm in the process of trying to automate the end-phase of my photo
workflow. That is, the portion after I've pulled the image into
Photoshop [Elements] or something and gotten it cropped and adjusted
"just so".
And that's gotten me to thinking relatively deeply about things that may
not yet admit deep thinking. So, I'm going to throw some ideas out to
the "PDML At Large" and see what comes back.
I want to provide myself some recourse against unauthorized
reproduction. So I'm thinking about several technologies and processes
to do that. In the past, my "attempts" have largely centered around
providing images on the web that are too small (in resolution terms) to
be of any real use outside the "Webisphere", and not caring about uses
within that environment.
I'm still planning to pursue that same strategy for the stuff I make
available to the world at large. Nothing more than, say, 800x600
pixels, for example. Put my copyright in the EXIF/IPTC metadata.
"Brand" a watermark visually into the image. Stuff like that.
But now it's looking like I can actually make some money from at least
some of my photographic endeavors. So, I want to afford myself some
more, not really protection, but recourse, ass coverage, whatever.
Increase my ability to "prove" that I originally created an image, after
that image has been cropped, resized, and otherwise mangled.
So I'm thinking about several aspects of deterring unauthorized use.
Phase one is to "brand" the images with a low-contrast modification that
imposes a notice visibly on the image, and keep the published resolution
"impractically" low. So, just how "visible" is too much in a watermark?
We're only talking about an 800x600 image, after all. Does anyone
have any experience with the "pay-to-play" image watermarking services?
I'm also thinking about embedding additional data via steganography.
Does anyone have any pointers or information about creating a "proper"
stegano-embeddable image that's not going to either detract from the
top-level image or be "too detectable"? Should I think about multiple
stegs, with a different data set each time? Just how resilient are
stegs in the face of image modifications like crops, resizes, and
replacement of the "brand"?
I'm also thinking about "shaving" every published image so that none of
them have the outside 1-10% of the image. Theoretically, if I suspect
unauthorized use, this should help me prove original ownership. But,
when do I shave them? Immediately after capture? Immediately prior to
publishing a particular rendering? Several times in the middle of the
workflow?
Discuss ...
:-)
--
Thanks,
DougF (KG4LMZ)
--
PDML Pentax-Discuss Mail List
[email protected]
http://pdml.net/mailman/listinfo/pdml_pdml.net
to UNSUBSCRIBE from the PDML, please visit the link directly above and follow
the directions.
