Posted by diverD on May 23, 2005 at 21:15:52:
In Reply to: I'm a looking, but??? (-) posted by seahunt on May 22, 2005 at 19:45:55:
Okay I downloaded them and Iím guessing they all began as Digital Camera Tiffs ...
...as opposed to digital camera JPEGs. Now Iíll start qualifying. There are several ways to identify with 100% certainty the difference between digital camera shots and scanned film. Conversely, it would not be difficult for a practiced Photoshop user to disguise some of these dead giveaways if they wonted to play ďStump the Experts.Ē
Dead giveaway #1 is film grain. Even my tiny, $200 Scamaker 6000, scanning a slide that is still mounted, not taped down, with no oil for a focus plane still produces evidence of film grain. Digital camera CCDs also have grain but they look totally different. But a bit of clever resizing followed by JPGing can cover up the film grain or at least make it look more like a CCDs noise which is more squarish than round.
Dead Giveaway #2 is the noise Channels. The noise channel is generally the channel with the least amount of detail. With RGB this is the Blue channel, with CMYK it is the Yellow channel. I doubt any one here scans directly to CMYK so lets just skip that. A digital camera image captured in JPG shows a very distinct jpg compression pattern in all channels especially the noise channel compared to a scanned image saved as a Tiff. But again, an image captured by a digital camera still has a more JPG like noise in the blue channel than a scanned image would show.
Dead giveaway #3 is the pixilation. A scanned slide from my inexpensive Scanmaker, scanned at the slide appropriate resolution of 3600 dpi at 100% will produce an image of about 8x6 at 300 dpi, this means with no resampling. You would have to zoom in for days to see any evidence of pixilation. this is not going to happen with a digital camera image. One of level of zoom will show pixilation right away. But again, a very simple thing for someone even half way clever to disquise. Just resample to 72 dpi and you get pixelation city.
Dead giveaway #4 (finally) is image proportion. Rogers images are all 640 x 480 pixels. Iím sure everyone with a digital camera is familiar this 640 x 480 dimmension. Film does not share the same aspect ratio as CCDs and scanners really donít have an aspect ratio. Again, it would have been ingenious to take scanned images of slides, crop them to 640 x 480 and see if you could fool someone. But then again, how many people out there are that ingenious. (Besides me, of course! HaHa :)
Therefor, based on all of the above, but especially by the 640 x 480 aspect ratio, Iím guessing all of Rogerís images originated as digital camera photos captured in Tiff mode.
Whew! What a lot off topic imaging blab. But since Iím replying to seahunt, Iím hoping Iíll get away with it!
I hope you enjoyed my dissertation. When on a dive boat, Roger once helped me with some advice on diffusing the daylights on my non-TTL manual strobe because I didnít like how hot my photos were turning out. It was great advice. I hope something Iíve written here will help someone else out.
Post a Followup