Photoshopped or not?

General UFO stories

Moderators: ryguy, chrLz, Zep Tepi

Re: Photoshopped or not?

Postby elevenaugust » Sat Dec 11, 2010 10:26 pm

Edit to add:

Searching for clues all other the net, seems like a guy from Amkon (Cogburn) succeed in obtained, not the original, but at least a better version of the photography directly with Sacha Christie.

What is interesting is that he passed it through JPEGSnoop and was able to compare the compression algorithms to cameras samples:

"Though there is no EXIF data in the images provided. However, I was able to match the compression used to a Sony Cybershot U (DSC-U20) camera, although secondary analysis was unable to confirm this, indicating the image was post-processed.

Furthermore, the images provided were also sampled at 70%-75% of the original image, indicating that these images were processed. There were some data fragments that suggested that the images were manipulated with Photoshop v 7.0, though this was not conclusive."


So, the camera used was most likely a DSC-U20.

Here's what steves-digicam says about this camera:

Image
IPACO, the new tool for photo and video analysis is on-line! www.ipaco.fr
User avatar
elevenaugust
 
Posts: 30
Joined: Wed Jul 07, 2010 8:55 pm


Re: Photoshopped or not?

Postby Slippery Jim » Sun Dec 12, 2010 1:27 am

Lets just admit it wasn't an anomalous phenomena. :wink:
You couldn't find your own arse with two hands, a map and Google!
User avatar
Slippery Jim
On A Quest for Reality
On A Quest for Reality
 
Posts: 161
Joined: Sun Nov 14, 2010 6:14 am

Re: Photoshopped or not?

Postby Access Denied » Sun Dec 12, 2010 4:33 am

elevenaugust wrote:Searching for clues all other the net, seems like a guy from Amkon (Cogburn) succeed in obtained, not the original, but at least a better version of the photography directly with Sacha Christie.

What is interesting is that he passed it through JPEGSnoop and was able to compare the compression algorithms to cameras samples:

Considering the source I would be highly skeptical of this claim...

Does that make any sense to anybody?

By the way, I found your numbered and enhanced picture interesting. I didn't notice this before but it looks like maybe both of the younger two are holding cell phones... was everybody there taking pictures? ;)

At any rate, it’s ridiculous that the original image with the EXIF data intact is being withheld… aren't these same people demanding disclosure of the “truth” about Rendlescam? :roll:
Men go and come but Earth abides.
User avatar
Access Denied
1 of the RU3
 
Posts: 2740
Joined: Tue Jan 09, 2007 7:32 am
Location: [redacted]

Re: Photoshopped or not?

Postby chrLz » Sun Dec 12, 2010 7:39 am

Access Denied wrote:
elevenaugust wrote:Searching for clues all other the net, seems like a guy from Amkon (Cogburn) succeed in obtained, not the original, but at least a better version of the photography directly with Sacha Christie. What is interesting is that he passed it through JPEGSnoop and was able to compare the compression algorithms to cameras samples:

Considering the source I would be highly skeptical of this claim...

Dunno anything about the Amkon and Cogburn names, but I can tell you that, yes, many cameras do have recognisable signatures in their compression algorithms, but it is not a 100% proof and there are various ways to fake it - plus I've seen JPEGSnoop misidentify one of my own images.

Also, the online services that claim to be able to pick 'photoshopping'.. again, yes, there can be traces left by image editors and yes that can be enhanced by software.. but equally, such techniques are not foolproof and there are many provisos. If someone just rolls out one of these tools and claims it proves such and such, they probably haven't read those provisos..

The reality is that even if you have the original camera media in hand and you saw the image taken and then watched the card be removed from the camera and handed to you, fakery can still be done in ways that are difficult or impossible to detect.

That's why the full story that comes with the image (or more commonly doesn't) is so important to establish provenance. Beyond that, the best 'analysis' comes from those who have seen an awful lot of images, taken an awful lot of images, and understand the imaging process, from scene to lens to post-proc...

EXIF is nice, but EXperience is nicer...


BTW, unless the camera is VERY broken, there is no way that the 'pre-flash' (used for red-eye, exposure or even focus) will appear in one of its own images. Plus the direction of the lighting is clearly different - so it's not a double flash from the same camera.
"To wear the mantle of Galileo, it is not enough that you be persecuted by an unkind establishment. You must also be right." - Robert L. Park (..almost)
User avatar
chrLz
Moderator
 
Posts: 258
Joined: Tue Nov 09, 2010 9:47 am

Re: Photoshopped or not?

Postby elevenaugust » Sun Dec 12, 2010 8:36 am

chrLz wrote:Dunno anything about the Amkon and Cogburn names, but I can tell you that, yes, many cameras do have recognisable signatures in their compression algorithms, but it is not a 100% proof and there are various ways to fake it - plus I've seen JPEGSnoop misidentify one of my own images.

Also, the online services that claim to be able to pick 'photoshopping'.. again, yes, there can be traces left by image editors and yes that can be enhanced by software.. but equally, such techniques are not foolproof and there are many provisos. If someone just rolls out one of these tools and claims it proves such and such, they probably haven't read those provisos..

The reality is that even if you have the original camera media in hand and you saw the image taken and then watched the card be removed from the camera and handed to you, fakery can still be done in ways that are difficult or impossible to detect.

That's why the full story that comes with the image (or more commonly doesn't) is so important to establish provenance. Beyond that, the best 'analysis' comes from those who have seen an awful lot of images, taken an awful lot of images, and understand the imaging process, from scene to lens to post-proc...

EXIF is nice, but EXperience is nicer...

Yes, definitely agree with all the above.

Using for example an hexadecimal editor can easily be done to fake all the EXIFs datas, however, amongst all the faked photos I've seen all around the net since years, it occurred only one time for my experiences and, yes, only a full background investigation in this case was able to expose the faker.
So, for someone to have the skills to fake a photo without letting any clue behind him is not an everyday job...

chrLz wrote:BTW, unless the camera is VERY broken, there is no way that the 'pre-flash' (used for red-eye, exposure or even focus) will appear in one of its own images. Plus the direction of the lighting is clearly different - so it's not a double flash from the same camera.

Not sure about this as the shadows are not that evident to see in this poor resolution photography, anyway, taking photography using double flash can be done, as I showed it above.
Is it possible to consider the use of a pre-flash + a flash as a double flash? :wink:

Sometimes, strange things occurs with pre-flash, look at this over-exposed photography:

Image

With the guy comments:
"A strange thing happened while shooting at a nightclub with my Canon 5D mark II. This happened twice in the 40 or so pictures that I've taken in very dark nightclubs. Both times, I saw people posing to have their picture taken, so I pulled up my camera and held the shutter button all the way down, so that it would shoot as soon as it was in focus. The pre-flash on the other cameras must have helped my camera focus. Most cameras with pre-flash (for red-eye reduction or to help focus) flash about 3 times within ½ second, and then the real flash goes off about ½ second later. I don’t know if this exposure was made during a pre-flash, or the real flash, but that amazing timing."

SOURCE

The two cameras theory is very interesting, but I stand with a simpler explanation as an alternative possible one...

Anyway, you're all right, without the original photography, any further work is useless... :roll:
Last edited by elevenaugust on Sun Dec 12, 2010 10:02 am, edited 1 time in total.
IPACO, the new tool for photo and video analysis is on-line! www.ipaco.fr
User avatar
elevenaugust
 
Posts: 30
Joined: Wed Jul 07, 2010 8:55 pm

Re: Photoshopped or not?

Postby elevenaugust » Sun Dec 12, 2010 8:37 am

Slippery Jim wrote:Lets just admit it wasn't an anomalous phenomena. :wink:

Nothing paranormal here, that's for sure...
IPACO, the new tool for photo and video analysis is on-line! www.ipaco.fr
User avatar
elevenaugust
 
Posts: 30
Joined: Wed Jul 07, 2010 8:55 pm

Re: Photoshopped or not?

Postby Access Denied » Sun Dec 12, 2010 6:04 pm

chrLz wrote:Dunno anything about the Amkon and Cogburn names, but I can tell you that, yes, many cameras do have recognisable signatures in their compression algorithms, but it is not a 100% proof and there are various ways to fake it - plus I've seen JPEGSnoop misidentify one of my own images.

Right, I did a little research on this…

Using JPEG quantization tables to identify imagery processed by software
Jesse D. Kornblum
Defense Cyber Crime Institute, United States
http://www.dfrws.org/2008/proceedings/p21-kornblum.pdf

The quantization tables used for JPEG compression can also be used to help separate images that have been processed by software from those that have not. This loose classification is sufficient to greatly reduce the number of images an examiner must consider during an investigation. As illicit imagery prosecutions depend on the authenticity of the images involved, this capability is an advantage for forensic examiners. This paper explains how quantization tables work, how they can be used for image source identification, and the implications for computer forensics.

And there appears to some potential problems with JPEGsnoop…

Ideally, the examiner could use the JPEG quantization tables to determine exactly what kind of device created each image and categorize the images accordingly. The program JPEGsnoop aims to do exactly this (Hass, 2008). The program comes with a database of tables that be compared against input files. Unfortunately, however, JPEGsnoop assumes that each camera can use only one quantization table. The use of custom adaptive tables, however, means that programs like JPEGsnoop would need to hold an unwieldy number of tables to be practical. Worse, some tables may be used by several devices, including both cameras and software programs, rendering the database inaccurate when attempting to determining an image’s origin.

Indeed that assumption appears to be the case…

http://www.impulseadventure.com/photo/j ... ation.html

But that wasn’t really what I was asking about. :)

Does it make sense given, according to Cogburn, the image was “sampled at 70%-75% of the original image” for it to retain the JPEG compression signature of the camera? Seems to me it would then have the signature of the program used to resize it…
Men go and come but Earth abides.
User avatar
Access Denied
1 of the RU3
 
Posts: 2740
Joined: Tue Jan 09, 2007 7:32 am
Location: [redacted]

Re: Photoshopped or not?

Postby elevenaugust » Sun Dec 12, 2010 7:53 pm

Great research "Acces denied".

I use JPEGSnoop in photographic analysis, but never alone, as all the above are known problems...
There are other possibilities to study in a forensic way a photography, such as shadows discrepancies, barrel/pincushion distortion amount, size/resolution, etc...

Anyway, in most cases, JPEGSnoop can be useful as 80-90% of the faker aren't aware of these problems and don't ever know how to fake EXIFs...

Does it make sense given, according to Cogburn, the image was “sampled at 70%-75% of the original image” for it to retain the JPEG compression signature of the camera? Seems to me it would then have the signature of the program used to resize it…

No, it doesn't make sense at all. Any useful forensic work using JPEGSnoop can only be done with the original.
Otherwise, it will of course give you only an indication of the post-process software that was used.
Moreover, seems like "Cogburn" said that the "original" photography that was sent to him by Sacha Christie had no EXIFS datas at all... :roll:
IPACO, the new tool for photo and video analysis is on-line! www.ipaco.fr
User avatar
elevenaugust
 
Posts: 30
Joined: Wed Jul 07, 2010 8:55 pm

Re: Photoshopped or not?

Postby chrLz » Mon Dec 13, 2010 2:06 pm

Access Denied wrote:...
Does it make sense given, according to Cogburn, the image was “sampled at 70%-75% of the original image” for it to retain the JPEG compression signature of the camera? Seems to me it would then have the signature of the program used to resize it…


What 11/08 said... :) Once you resample, jpeg compression artefacts (and any others) may of course be seen by the resampling algorithm as 'real' data, 8/16-pixel q-block boundaries are replaced/overwritten with new ones, the algorithm makes its own mathematical guesses, and kazam.
So any traces that were already difficult if not impossible for useful identification, have now been, well.. 'decimated'.. (to use a very obscurely apt signal-processing-related pun..)
"To wear the mantle of Galileo, it is not enough that you be persecuted by an unkind establishment. You must also be right." - Robert L. Park (..almost)
User avatar
chrLz
Moderator
 
Posts: 258
Joined: Tue Nov 09, 2010 9:47 am

Re: Photoshopped or not?

Postby elevenaugust » Mon Feb 07, 2011 10:29 pm

Yeah, old topic, I know but couldn't resist... :D

A similar sample:

Image

I'm in the process getting more informations from the photographer, including the original photographies, if there was any other people with camera around there, etc etc...

Will let you know. :wink:
IPACO, the new tool for photo and video analysis is on-line! www.ipaco.fr
User avatar
elevenaugust
 
Posts: 30
Joined: Wed Jul 07, 2010 8:55 pm

Re: Photoshopped or not?

Postby RICH-ENGLAND » Tue Feb 08, 2011 7:54 pm

@ elevenaugust

nice find, is there also a "paranormal" story attached to that picture ?.

thanks

rich
ATS HAS TURNED INTO A "BALLOONATIC" ASYLUM
User avatar
RICH-ENGLAND
Focused on Reality
Focused on Reality
 
Posts: 343
Joined: Mon Nov 29, 2010 4:06 am

Re: Photoshopped or not?

Postby elevenaugust » Tue Feb 08, 2011 9:30 pm

Well, not really, but this photography was taken at a known place in France for alleged paranormal and UFOs activities: the "Col de Vence".

There were lots of photographies taken over there that are most likely camera or flashes artifacts, etc...
IPACO, the new tool for photo and video analysis is on-line! www.ipaco.fr
User avatar
elevenaugust
 
Posts: 30
Joined: Wed Jul 07, 2010 8:55 pm

Re: Photoshopped or not?

Postby Access Denied » Wed Feb 09, 2011 6:57 am

Like Rich said, nice find. Judging by the direction and sharp definition of the shadow to the right, looks like there was another flash or bright source of light just to the left. If you’re able to get some more information it would be interesting to see the EXIF data to get the exposure time and find out what kind of camera it was… although if it was malfunctioning for some reason, I would be suprised if there was any indication of that.
Men go and come but Earth abides.
User avatar
Access Denied
1 of the RU3
 
Posts: 2740
Joined: Tue Jan 09, 2007 7:32 am
Location: [redacted]

Re: Photoshopped or not?

Postby RICH-ENGLAND » Wed Feb 09, 2011 1:55 pm

elevenaugust wrote:Yeah, old topic, I know but couldn't resist...


well i'm glad you revived it as i would like to know how it was done, as even the resident photography experts are not certain. so im sure i can learn a lot from the discussion between you guys that know about this stuff.

thanks

rich
ATS HAS TURNED INTO A "BALLOONATIC" ASYLUM
User avatar
RICH-ENGLAND
Focused on Reality
Focused on Reality
 
Posts: 343
Joined: Mon Nov 29, 2010 4:06 am

Previous

Google

Return to UFOs

Who is online

Users browsing this forum: Bing [Bot] and 17 guests

cron