Computers and geometry can unmask fake photos

U.S. researchers say new software can detect faked or altered photos using a geometric algorithm to locate inconsistent shadows not obvious to the naked eye. Computer scientists at Dartmouth College and the University of California, Berkeley, said the development could be significant step in the field of digital forensics, allowing investigators to differentiate between authentic images and computerized forgeries. The new method analyzes a variety of shadows in an image to determine if they are physically consistent with a single illuminating light source, allowing a forensic analyst to determine if a photo is physically plausible or the result of image fakery, a Dartmouth release said Monday. "Our method shifts the dialogue from 'does the lighting/shadow look correct?,' which is well known to be highly unreliable, to a discussion of whether an analyst has correctly selected the location of cast and attached shadows in an image, a far more objective task," Dartmouth computer scientist Hany Farid said. In one example, the method has been used to debunk claims the lighting and shadows in the famous 1969 moon landing photo are fake, the researchers said. "In this regard, our method lets humans do what computers are poor at -- understanding scene content -- and lets the computer do what humans are poor at -- assessing the validity of geometric constraints," Farid said.