Watching trial. Defense and prosecution are arguing whether the State’s pixel-added though interpolation alters the image evidence. Specifically, defense is concerned about added/changed colors.
Valid argument. Personally, I think enhanced/interpolated images should only be as an investigative tool, to find evidence. It should not be used as evidence at trial.
But in this case,, I wish someone were knowledgeable to ask for comparative histograms from the original and interpolated images. That would show any significant change in tonal distribution.
ADDED: BTW, the State’s imagery dude just admitted that he used frame averaging to CREATE a new, never before existent image to introduce as trial evidence.
I repeat: Investigative tool, not evidence.
Explanation: Take five sequential old-fashioned projector slides from high speed film. Instead of looking at them one at a time, stack them together. You now get a conglomerate blob of some object. Now let your computer average pixels; add some, subtract some. You might get a better picture of an apple that was there, or you might get an apple that was never there before. The picture is not real.
If you’re doing art, fine.
If you’re helping a detective figure out if he should look for evidence of an apple (or gun) on the scene, fine.
If you take it into court and tell the judge and jury that it is an apple (or gun), YOU FALSIFIED EVIDENCE. Do not pass go. Go directly to jail.
|If you found this post useful, please consider dropping something in my tip jar. I could really use the money, what with new cell phone, ISP bills, SSL certificate, and general life expenses.Click here to donate via PayPal.|