Photography and forensic science

Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

On 10-02-07 17:13 , Charles wrote:
Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

There is no reason that a raw cannot be faked from a high res JPG or
TIF. However it won't map back exactly as the original. How much
weight from given pixels to 'de-mosaic' into the RGB "sets" is plain
guess work.

--
gmail originated posts are filtered due to spam.

On 2/7/2010 2:40 PM, Alan Browne wrote:
On 10-02-07 17:13 , Charles wrote:
Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

There is no reason that a raw cannot be faked from a high res JPG or
TIF. However it won't map back exactly as the original. How much weight
from given pixels to 'de-mosaic' into the RGB "sets" is plain guess work.

Right, the raw files are a grid of red, green and blue pixels, when
demosaiced, the green ones (twice as many) are given priority for
lightness/darkness and the others are spread around to average things
out. I don't know how you'd ever figure out how to un-average them back
to the exact same places given each pixel has contributions from
probably at least 9 adjacent pixels. You could make a raw file but not
the same raw file. So you could make a totally fake picture but not
modify a picture where authorities had a copy of the original raw.

On 2010-02-07 14:13:31 -0800, "Charles" said:

Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

Some Nikon DSLRs have a forensic marker that can be put in their NEF
files that makes it obvious if the file has been changed in any way.
People may be thinking of that. Nikon has a subsidiary called Nikon
Forensic Services which trains people in forensic photography and
performs other services.

Otherwise, what Alan said.

--
Waddling Eagle
World Famous Flight Instructor

C J Campbell wrote:
"Charles" said:

Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

Some Nikon DSLRs have a forensic marker that can be put in their NEF
files that makes it obvious if the file has been changed in any way.

Ditto Canon. It's useful for police departments.

--
Ray Fischer

On 08/02/2010 07:08, Ray Fischer wrote:
C J wrote:
said:

Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

Some Nikon DSLRs have a forensic marker that can be put in their NEF
files that makes it obvious if the file has been changed in any way.

Ditto Canon. It's useful for police departments.


How does one detect that one is taking a picture of a printout of a
doctored picture?

--
Bertrand

"Charles" wrote:

Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

Not all mathematical manipulations are reversible. Consider simple addition. If
the answer is 5, you don't know whether it was 1+4 or 2+3. -- Doug

On 10-02-08 10:30 , Douglas Johnson wrote:
wrote:

Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

Not all mathematical manipulations are reversible. Consider simple addition. If
the answer is 5, you don't know whether it was 1+4 or 2+3. -- Doug

Perfectly Reversible is not required. An astute estimate of the
weightings from 3 color pixels to single color pixels in a small area is
all that is needed (combined with appropriate weights of those colours
from adjacent pixels).

Mathematics is replete with estimation techniques where perfect
solutions do not exist or where they complex or computationally intensive.

--
gmail originated posts are filtered due to spam.

Alan Browne wrote:

On 10-02-08 10:30 , Douglas Johnson wrote:
wrote:

Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

Not all mathematical manipulations are reversible. Consider simple addition. If
the answer is 5, you don't know whether it was 1+4 or 2+3. -- Doug

Perfectly Reversible is not required.

No, of course not. I was addressing the narrow point of "mathematical
manipulations that surely are reversible." They most likely are not. However,
as you point out, they don't have to be.

Even the article says:

"Whatever demosaicing algorithm is applied, the pixels in the final digital
image will be correlated with their neighbors. If an image does not have the
proper pixel correlations for the camera allegedly used to take the picture, the
image has been retouched in some fashion."

So RAW files can probably be faked. You just have to be careful about it.

-- Doug

On 10-02-09 10:39 , Douglas Johnson wrote:
Alan wrote:

On 10-02-08 10:30 , Douglas Johnson wrote:
wrote:

Ran across some interesting posts about faked photos. Led me to this:

http://www.scientificamerican.com/ar...to-spot-a-fake

Then, ran across some opinions that RAW files cannot be faked? Wait a
minute! The demosaicing algorithms are public knowledge. Algorithms are
simply mathematical manipulations that surely are reversible. What am I
missing here?

Not all mathematical manipulations are reversible. Consider simple addition. If
the answer is 5, you don't know whether it was 1+4 or 2+3. -- Doug

Perfectly Reversible is not required.

No, of course not. I was addressing the narrow point of "mathematical
manipulations that surely are reversible." They most likely are not. However,
as you point out, they don't have to be.

Your entire message (above) implied there was no way out.

--
gmail originated posts are filtered due to spam.