Hany Farid from Darmouth published a tutorial on image forensics. Some image sensor artifacts help to recognize a forgery:
"a CFA interpolated image can be detected (in the absence of noise) by noticing, for example, that every other sample in every other row or column is perfectly correlated to its neighbors. At the same time, the non-interpolated samples are less likely to be correlated in precisely the same manner. Furthermore, it is likely that tampering will destroy these correlations, or that the splicing together of two images from different cameras will create inconsistent correlations across the composite image. As such, the lack of correlations produced by CFA interpolation can be used to expose it as a forgery."
Also, PRNU can be used to determine if an image is originated from a specific camera, or if a portion of an image has been altered.
Maybe Column FPN in CMOS sensor can give better clue ...
ReplyDeletewhat happens to these "fingerprints" if one properly flat-fields the images?
ReplyDeleteit seems that if the flat fielding is done correctly that the FPN components would all be removed.
If the image was then down sampled 2x2, would that cause most of the color deinterlace/interpolation artifacts to be removed?
Always good to read on a blog from halfway around the globe what my colleague across the quad is up to! ha!
ReplyDelete