Beta
×

Welcome to the Slashdot Beta site -- learn more here. Use the link in the footer or click here to return to the Classic version of Slashdot.

Thank you!

Before you choose to head back to the Classic look of the site, we'd appreciate it if you share your thoughts on the Beta; your feedback is what drives our ongoing development.

Beta is different and we value you taking the time to try it out. Please take a look at the changes we've made in Beta and  learn more about it. Thanks for reading, and for making the site better!

Nikon's Image Authentication Insecure

timothy posted more than 3 years ago | from the pictures-don't-lie-cameras-lie dept.

Encryption 106

silanea writes "Elcomsoft claims to have broken Nikon's Image Authentication system which — apparently only in theory — ensures that a photograph is authentic and not tampered with through a digital signature. They were able to extract the signing key from a camera and use it to have a modified image pass the software verification, rendering the rather expensive feature mostly marketed to law enforcement all but useless. So far Nikon has not given a statement. Canon's competing system was cracked by the same company last December."

cancel ×

106 comments

Sorry! There are no comments related to the filter you selected.

This is great news (3, Funny)

gnick (1211984) | more than 3 years ago | (#35969218)

Whew - I've always hated having to wear a ski mask when I "work". Now I can just claim image tampering.

Right (1)

elsJake (1129889) | more than 3 years ago | (#35969228)

Not like anybody would've expected that ...no way ...

Re:Right (1)

RDW (41497) | more than 3 years ago | (#35974780)

'Not like anybody would've expected that ...no way ...'

Especially given Nikon's less than stellar record with encrypting stuff previously. Remember the fuss about Nikon's white balance encryption a few years ago?:

http://it.slashdot.org/story/05/04/25/0511241/Adobe-Blasts-Nikons-Closed-File-Format [slashdot.org]

Adobe cut a deal with Nikon over this to avoid potential DMCA violations (Adobe Camera Raw uses some Nikon code to decrypt white balance), but everyone else just reverse-engineered Nikon's 'secret' key, which turned out not to be much of a challenge - it's now the 'xlat' table in the dcraw and ExifTool source:

http://cpan.uwinnipeg.ca/htdocs/Image-ExifTool/Image/ExifTool/Nikon.pm.html [uwinnipeg.ca]

(Nikon obfuscates things somewhat by also using the camera serial number and shutter count as keys).

From the Elcomsoft article on the latest crack:

'Two 1024-bit (128-byte) signatures are stored in EXIF MakerNote tag 0Ã--0097 (Color Balance).'

This is the same tag that Nikon still uses to store white balance values encrypted with their broken xlat key (which dcraw, ExifTool and others routinely decrypt). Of course the difference here is that image authentication was a feature designed for the benefit of the (forensic) user, whereas white balance encryption was intended to benefit only Nikon by denying third party software access to important metadata. But both are now equally broken.

Great (0, Offtopic)

Anonymous Coward | more than 3 years ago | (#35969242)

With this secure authentication feature, we will finally know which of the photos in the 'Celebrities' sections of porn sites are real!

The danger of these systems is they appear secure (5, Insightful)

SuperKendall (25149) | more than 3 years ago | (#35969352)

This is great news, because now people will be able to cast doubt on images when there is cause to instead of being told "it's not possible it's a fake, it's signed". You know that if someone cracked it publicly someone else (probably many someone else's) have cracked it in private, and have kept around the ability to forge photographs in case of emergency... that ability is now reduced.

Re:The danger of these systems is they appear secu (4, Interesting)

roc97007 (608802) | more than 3 years ago | (#35969718)

I think the authorities will still say "it's not possible it's a fake, it's signed" and it'll be up to the victim (or the victim's lawyer) to know that the signage has been broken.

The last time I was stopped in a speed trap (on motorcycle), I knew it was coming up (they always put a speed trap in this particular construction zone on weekends because people ignore the temporary "35" signs 'cause there's nobody working on Sunday, but I digress) and had slowed way down before taking the turn, but was waved over anyway. I was pretty sure he'd tracked the (obviously faster) car one lane over instead of me, and said so. He said "the gun can't be wrong, I had a firm lock on you." I can see the stupid radar gun in his hand right there, and it's not like there's a scope on it, or even if he actually had me in crosshairs, that it could tell the difference between a slow moving object in the foreground and a much faster object in the background. I maintained that he could not possibly have locked on me, because he would have read 33 MPH, which is what my speedo was displaying at the time. I said it obviously had "locked" on the car that passed me shortly after the corner. The cop said that this was impossible, radar guns don't make that kind of mistake.

Well hell, there's a huge body of evidence that radar guns make "mistakes" all the time. I laid out exactly how the error could have occurred, he continued to insist that the gun can't make mistakes. I finally said "ok, whatever. We'll see what the judge says." He went away, talked to his cohorts for awhile, came back and issued me a "verbal warning", let me go. Now, I strongly suspect that if I'd acted like I knew nothing about the technical details of radar guns, I'd have gotten a ticket.

Re:The danger of these systems is they appear secu (1)

Anonymous Coward | more than 3 years ago | (#35970618)

Good on you for standing up for yourself! I know that police like to use Gatso 24 AUS-series doppler radar units around the 40kmh school zones in Australia.

The problem with this is that the Gatso operations manuals as well as the police operations guidelines say that these units are not to be used in zones signed less that 60kmh (they're unreliable at low speeds) except that nobody in the public would know this or even consider questioning them over it. Most people wouldn't even bother to ask the officer what they've set the cosine correction factor to, or ask to see the calibration report that they must fill out at every location they setup (every time they setup).

Re:The danger of these systems is they appear secu (1)

davester666 (731373) | more than 3 years ago | (#35971704)

I don't understand why you weren't tazered?

Re:The danger of these systems is they appear secu (1)

roc97007 (608802) | more than 3 years ago | (#35971884)

It was a few years ago...

Re:The danger of these systems is they appear secu (1)

Ginger Unicorn (952287) | more than 3 years ago | (#35972926)

Question marks are for questions?

Re:The danger of these systems is they appear secu (1)

Vegemeister (1259976) | more than 3 years ago | (#35975818)

Yes. Like the one you just asked.

Re:The danger of these systems is they appear secu (0)

Anonymous Coward | more than 3 years ago | (#35971732)

You're lucky he didn't give you one just for being sure of yourself / mouthy / "trying to get out of a ticket".

Re:The danger of these systems is they appear secu (1)

roc97007 (608802) | more than 3 years ago | (#35971888)

I wasn't "trying to get out of a ticket". I wasn't guilty of the infraction of which I was accused. Have we really gotten to the point where professing innocence is a sign of guilt?

Re:The danger of these systems is they appear secu (1)

pnutjam (523990) | more than 3 years ago | (#35974744)

only the last part of your argument was correct, "let see what the judge says".

The racially ambiguous (not quite white, if you prefer) do not argue with cops...

Re:The danger of these systems is they appear secu (0)

Anonymous Coward | more than 3 years ago | (#35975158)

You are my hero.

Re:The danger of these systems is they appear secu (1)

rolfwind (528248) | more than 3 years ago | (#35973012)

This is great news, because now people will be able to cast doubt on images when there is cause to instead of being told "it's not possible it's a fake, it's signed". You know that if someone cracked it publicly someone else (probably many someone else's) have cracked it in private, and have kept around the ability to forge photographs in case of emergency... that ability is now reduced.

And yet corporations the world over are clamoring or have made this type of hacking, even on your own bought stuff, illegal.

Glitch in the Matrix? (1)

easyTree (1042254) | more than 3 years ago | (#35969380)

I'm getting a serious case of déjà vu...

Easy to fake (5, Insightful)

jdbannon (1620995) | more than 3 years ago | (#35969392)

Just take a picture of the photo-shopped image with your Nikon camera. Bam! That was sure hard to crack.

Re:Easy to fake (2)

Gordo_1 (256312) | more than 3 years ago | (#35969892)

Ah yes, the ever present analog loophole. How soon before the camera manufacturers come up with a technology that prevents the digital signature from being applied to a picture when a large 2-dimensional plane parallel to the sensor is detected? And how long before some Julian Beever [wikipedia.org] wannabe finds a way around that?

Re:Easy to fake (1)

ThatsMyNick (2004126) | more than 3 years ago | (#35970104)

when a large 2-dimensional plane parallel to the sensor is detected?

You mean like a wall?

Re:Easy to fake (1)

petermgreen (876956) | more than 3 years ago | (#35970184)

What I suspect he means is a device placed directly on the sensor to allow pixel perfect "fakes" to be applied to the sensor directly.

Re:Easy to fake (1)

ArundelCastle (1581543) | more than 3 years ago | (#35970350)

when a large 2-dimensional plane parallel to the sensor is detected?

You mean like a wall?

Yes, to photograph a wall authentically you'd need to purchase a special 2D camera with no depth of field.

Re:Easy to fake (1)

ProfessionalCookie (673314) | more than 3 years ago | (#35970806)

Or you need a not-flat wall.

Re:Easy to fake (1)

jimicus (737525) | more than 3 years ago | (#35972512)

I've already got one of those, it's called a macro lens.

Re:Easy to fake (1)

Alex Belits (437) | more than 3 years ago | (#35973278)

Yes! DRM'ed murals!

Re:Easy to fake (1)

Darinbob (1142669) | more than 3 years ago | (#35970904)

Then they'll jail you for breaking copyright!

Re:Easy to fake (2)

thegarbz (1787294) | more than 3 years ago | (#35971098)

I've heard this before, but how exactly do you propose to do this? Every image taken of a displayed medium be it paper under theoretical perfect lighting, or monitors with theoretical perfect backlight suffers quality issues that make it plainly bloody obvious that the picture was taken of a picture. There's no way around this for a few reasons.

Firstly the resolution of cameras will clearly show the defects in the material
Secondly the gamuts of printed paper and displays are smaller than those of the camera sensor and nature leading to muted colours
And Thirdly assuming we have a 10mpxl screen with a perfect backlight and perfect colour reproduction you'd still suffer a moire effect of the grid on the sensor not lining up with the grid on the display pixels.

Try this yourself. Take a photo of your monitor and have a look at just how god awful it turns out.

Re:Easy to fake (2)

jdbannon (1620995) | more than 3 years ago | (#35971956)

I think you could do a pretty good job with a semi-pro ten color inkjet. The gamut will be near sRGB. You can upscale the image and blur a bit to kill moire. Reducing the camera capture resolution and compression quality a touch would further hide any defects. Most importantly, if you tell a courtroom "Look, the picture looks good and Nikon cameras make magic pictures that can't lie." They are going to say "OK!" not "Why don't we analyze the image gamut and maybe look for double vignetting or warping that isn't quite consistent with a standard Nikon lens. Not only is this possible, it's how everything was actually done not too many years ago. The technique of "Cut up some printed images or film and lay them out to be photographed again" was the way that books, newspapers and magazines were created for many, many years. Do you think pre-1990s National Geographic pictures are too ugly and unbelievable to pass muster in a courtroom?

Re:Easy to fake (1)

thegarbz (1787294) | more than 3 years ago | (#35972398)

No I believe that any evidence that one party strongly believed is fake would go under intense scrutiny. There's always a chance people can take the image at face value. The clear indication in the above example is that the camera's dynamic range is waaay larger than that of even the best inkjet paper and under perfect conditions the image would look flat. (remember you can't touch up the picture in post to fix that one).

An expert would be able to tell. Heck post it on a photography forum and an amateur would be likely be able to tell. The point is it's easy to do a crap fake anywhere. Whether it would hold up in court though is a matter of luck.

Re:Easy to fake (1)

QuasiSteve (2042606) | more than 3 years ago | (#35975876)

The signing happens on a separate chip. This means that all you actually need to get a fictional piece of graphics to be signed is to fake being the device's sensor. This is well beyond the capabilities of the average Joe - but certainly not beyond technical capabilities.

The signing method itself being cracked, of course, puts the ability to get any image signed into the hands of said average Joe.

Re:Easy to fake (1, Interesting)

Anonymous Coward | more than 3 years ago | (#35971592)

And your signed exif data will show that the photo was taken with a macro lens focused 2 feet away.

Re:Easy to fake (2)

jdbannon (1620995) | more than 3 years ago | (#35971990)

This is a better objection. But as a rough plan, I'd put on a manual focus lens, and connect the circuitry to an auto-focus lens laid next to it. Tell the camera to focus into the distance, but focus your inline manual lens as you need to.

And... at this point it's easier probably to use the software crack. The point, though, is that next week there will be a new and "truly unbreakable" version of the software that closes whatever hole was found, but it sure seems like access to the hardware lets you defeat any possible system that could be designed.

Re:Easy to fake (1)

SharpFang (651121) | more than 3 years ago | (#35973512)

I guess the digital signature contains date&time. Of course this would assume camera's internal clock can't be tampered with.

Worth 1K Words (1)

Nom du Keyboard (633989) | more than 3 years ago | (#35969394)

So a picture is still worth a thousand words, but 999 of those words may be a lie.

Re:Worth 1K Words (1, Funny)

Archangel Michael (180766) | more than 3 years ago | (#35969486)

The picture is okay. It is the cake that is a lie!

release our perfect priceless progeny at once (0)

Anonymous Coward | more than 3 years ago | (#35969500)

or face what your comic style textbook deems, a very bad ending. at once mind you.

Nikon didn't learn from DRM (1)

NeutronCowboy (896098) | more than 3 years ago | (#35969508)

Where every local implementation of DRM has been broken. Sure, they could require a working internet connection for every picture taken, but I'm pretty sure even the laziest corporate-boot-loving shopaholic would draw the line at buying a camera with such a "feature".

Re:Nikon didn't learn from DRM (1)

owlstead (636356) | more than 3 years ago | (#35969684)

The internet connection would not work either: it would replace the security problems with the signing key with the security problems of the authentication key :)

Re:Nikon didn't learn from DRM (1)

0123456 (636235) | more than 3 years ago | (#35969744)

Basically it's impossible to do, because 'tamper-proof' hardware isn't. The only question is how hard you can make it.

And note that you don't actually have to get the key, if you could somehow hack into the feed from the CCD into the camera you could feed the fake picture in that way and have the camera sign it for you.

Re:Nikon didn't learn from DRM (2)

fuzzyfuzzyfungus (1223518) | more than 3 years ago | (#35969882)

As long as the signing key is unique per-camera(which I would bloody well hope it is, for forensic purposes), "tamper-evident" is arguably good enough, and probably easier to approach(as with any hardware security measure, the approach to the ideal is more or less asymptotic, with price spiking to near infinity as you reach the goal).

If the camera is tamper-evident, anybody who suspects manipulation of photos ostensibly from that camera can attack the credibility of the camera on technical grounds, just as they might a witness: "Your honor, the camera has probe traces on its 'secure' ROM pins, its private key could easily have signed more shit than John Hancock."

In such a case, anybody who wanted to use the camera for evidence gathering would be required to maintain physical security around it, as is necessary.

The problem crops up if the key can be extracted silently, or is shared between multiple cameras. Tamper-proof is optional. Tamper-evident is absolutely necessary, or doubt is cast on every image signed.

Re:Nikon didn't learn from DRM (1)

0123456 (636235) | more than 3 years ago | (#35970290)

If it's unique per camera, you'd need to be able to prove that key is in that camera.

Re:Nikon didn't learn from DRM (1)

Darinbob (1142669) | more than 3 years ago | (#35970994)

Just have it sign another picture to verify.

Re:Nikon didn't learn from DRM (2)

Zerth (26112) | more than 3 years ago | (#35974734)

And also prove you don't have a second camera that has been tampered with to have the same key as the untampered camera?

Re:Nikon didn't learn from DRM (0)

dgatwood (11270) | more than 3 years ago | (#35970310)

As others have mentioned there is no way to prove that a photo has not been doctored so long as it is possible to doctor it, compensate for lens distortion, print the photo out, and take a picture of the picture. Therefore, using this for tamper detection would inherently be prima facie worthless even if DRM weren't a fundamentally unsolvable problem.

That said, you're all missing the primary purpose of this image authentication. It's not to prove that a photo has not been doctored. It is to prove with a reasonable degree of certainty that the photo was taken with a particular camera. For kiddie porn, photos of crimes being committed, etc., that's a valuable piece of information.

Unless somebody caught producing such pictures can show reasonable cause to believe that somebody cracked his/her camera and stole the key, the mere existence of a verified signature is close enough to airtight to get a conviction. By contrast, previous methods (a serial number embedded in an EXIF tag) are too trivial to fake.

So you see, it's not about proving that an image has not been altered, but rather proving that the camera in question took the photo, providing evidence that its owner was almost certainly at the scene of the crime.

Re:Nikon didn't learn from DRM (1)

dgatwood (11270) | more than 3 years ago | (#35970336)

Let me correct that slightly. When it bubbles down into the lower end models, that's how it will be used. For now, it's a proof of concept.

Re:Nikon didn't learn from DRM (1)

fuzzyfuzzyfungus (1223518) | more than 3 years ago | (#35971446)

For that purpose, cryptographic signing would seem to be pretty useless: by design, the slightest modification to the file breaks the signature, and the signature is readily strippable. Signing is for people who want to prove that they did. Some variation of the various "fingerprinting" techniques that the Copy Cops have been trying on films and such would be what you would need to go after somebody who very much wants to hide their involvement in the chain.

In many cases, I suspect, the unique pattern of manufacturing irregularities in the CCD would probably be much harder to hide(without either throwing away massive amounts of image data, or using classy gear, multiple exposures, darkfield subtraction, etc.) than a cryptographic signature or some of the weaker fingerprinting techniques...

Re:Nikon didn't learn from DRM (1)

hairyfeet (841228) | more than 3 years ago | (#35970958)

Uhhh...sorry bud, but you're wrong. You see Nikon was selling this "feature' (at probably significant markup as with anything being bought by a government agent these days) to "prove" that the police camera hadn't been tampered with. So it wouldn't do squat in the case of some child pornographer who is probably using some CCC (cheapo Chinese crap) throwaway camera, since they wouldn't have the authentication and signing chip.

So this didn't have anything to do with catching the bad guy read handed, it was more to keep the defense lawyer from arguing the cop IS the bad guy, by doctoring evidence with his camera. Of course as other pointed out the analog hole made the whole thing an expensive waste anyway.

Re:Nikon didn't learn from DRM (1)

owlstead (636356) | more than 3 years ago | (#35970364)

It is unique per camera, it says so in the press release which I linked in a separate post. Unfortunately, I could not Google up any other details.

As in most problems regarding crypto, the usage scenario is rather important. For instance, I can imagine that editors of newspapers trust the photographer enough to keep the camera secure. In that case you can use the signature to verify that it was taken by the camera and photographer. That may not hold in court though, where the evidence is always suspect of deliberate falsification.

Of course, if the camera is tamper evident, a jurisdictional court is likely to have an expert look at it. Then again, if the problem is in the implementation of the signature then tamper evident does not mean squat. It seems we'll have to wait until the details become available.

Re:Nikon didn't learn from DRM (0)

Anonymous Coward | more than 3 years ago | (#35972206)

Even a taper evident camera can be reproduced faithfully by the manufacturer. So evidence produced by any party with enough leverage over the manufacturer can be suspected of being faked without that being evident from the camera.

They would just ask for two camera's with the same signing key. Or the combination signing key/camera.

Re:Nikon didn't learn from DRM (2)

RDW (41497) | more than 3 years ago | (#35975020)

'It is unique per camera, it says so in the press release which I linked in a separate post.'

I may be missing something, but I can't see this in the press release, so there may well just be a single key. However, every camera model with the image authentication feature also writes its (unencrypted) serial number to an EXIF tag. If image authentication had remained secure, you could have 'proved' which camera took the photo simply by reading the serial number from the metadata of an authenticated image (tampering with the number would invalidate the image).

Re:Nikon didn't learn from DRM (0)

Anonymous Coward | more than 3 years ago | (#35970988)

I have had to work on devices that automatically erase their keys if they detect debugger connections. Which is pretty annoying if you're a developer. Plus secure keys are the same chip as the cpu/memory/security-engine so you can't just probe the busses without cracking open chips.

If you're going to have a private key on a device, then you need to take the appropriate steps to protect it rather than rely on the fact that most people won't figure it out.

Re:Nikon didn't learn from DRM (1)

obarel (670863) | more than 3 years ago | (#35969708)

Well, they could require an internet connection for getting the pictures out of the camera.
Still, you'd have to place a very high value on authenticity (vs. convenience) to use it.

Possible algorithm:
1. Create a random symmetric key
2. Encrypt the picture using this key, and encrypt the key using the owner's public key
3. Create another random symmetric key
4. Encrypt the picture using (again) using this key, and encrypt the key using the company's public key

To extract:
1. You must send each picture to the company (along with the encrypted key)
2. The company decrypts the outer layer, signs the (encrypted) picture and sends it back
3. The inner layer is decrypted on the owner's computer (using her private key to reveal the inner key)

Job done - the company doesn't have access to the picture, yet the encrypted picture is signed. To prove that the picture is authentic - encrypt it using the random key and then verify the signature.

Re:Nikon didn't learn from DRM (1)

obarel (670863) | more than 3 years ago | (#35969810)

Replying to myself, but of course that wouldn't work...

The main problem is keeping the company's public key secure (otherwise I could just take any modified picture and follow the above). But that's impossible, as TFA proves...

Re:Nikon didn't learn from DRM (1)

MoonBuggy (611105) | more than 3 years ago | (#35969752)

Assuming that were a workable solution, I can see a market for it: as the summary mentions, these cameras are (unreasonably?) expensive and generally marketed at law enforcement - bundling a 3G modem and a service package to handle the authentication doesn't sound so onerous when the whole thing is required for legal compliance and goes down on the department budget.

That said, the assumption of workability is a higher bar than it might sound. Hackers are smart people, after all, and as another poster mentioned, the only thing worse than an insecure system is an insecure system that looks secure.

Re:Nikon didn't learn from DRM (1)

RDW (41497) | more than 3 years ago | (#35974110)

The cameras are actually standard models - all the recent pro and semi-pro models (D200 and up) support this feature, though it's off by default. Once you activate it in the camera's menu, any image you take can be 'authenticated' by the software, which goes for about $500 USD.

Re:Nikon didn't learn from DRM (2)

RightSaidFred99 (874576) | more than 3 years ago | (#35970008)

Yeah, just look at that thriving Xbox 360 pirate game environment.

With enough effort any DRM can be broken somehow. The only thing the content owner has to do is ensure that it's difficult and/or expensive enough to not be worth it.

And in case you didn't read the..summary, the camera is supposed to help provide a chain of evidence. Not sure why anyone would put "feature" in double quotes and act like nobody would buy a camera that supports this.

Re:Nikon didn't learn from DRM (1)

GameboyRMH (1153867) | more than 3 years ago | (#35973352)

That was my first thought, this is a stupid DRM-like idea. It's just a more complicated version of relying on a Word file's internal datestamps or a JPEG's EXIF metadata.

The only way an image auth system could work is if the camera had an always-on Internet connection that could send the time, the picture's hash, and the file size to a "hash server" (where you can't edit entries) and even the applications of that would be very limited - it could only prove that a picture existed at a certain time. If you fake up a Bigfoot photo on Tuesday and claim it was taken during your hike in the woods on Wednesday, and have a hash to back up that the photo existed in its current state on Wednesday, that doesn't prove anything.

Does it really matter? (1)

e9th (652576) | more than 3 years ago | (#35969512)

Has there ever been a case whose outcome depended on the authenticity of a digital image?

Re:Does it really matter? (0)

Anonymous Coward | more than 3 years ago | (#35969542)

There have been a number of cases of police editing video before turning it over to the defense. To date, the edits have been fairly simple -- deleting portions of the video which proved innocence.

Re:Does it really matter? (1)

mythosaz (572040) | more than 3 years ago | (#35969622)

I'm going to go with "lots." ...you know, like every one that involved a "photo" or "video" taken in the last decade.

Re:Does it really matter? (1)

e9th (652576) | more than 3 years ago | (#35969700)

I'm aware of all the 'shopped images out there. What I am looking for is a trial where one side or other claimed that an image in evidence was altered to affect the outcome.

Re:Does it really matter? (0)

Anonymous Coward | more than 3 years ago | (#35970234)

Don't know about any cases involving authenticity, but there have some trials where "computer enhanced" digital images have been admitted as evidence by some courts. For example, in STATE OF CONNECTICUT v. ALFRED SWINTON, the appeal court ruled that the lower court had not erred when it admitted into evidence a digitally enhanced image of a bite mark on a victim's breast. The image is clearly altered, but the court felt that the image wasn't altered in such a way as to present a false representation of fact. The appeal court did rule, however, that the lower court erred when it also admitted a photoshopped picture of the defendant's dental cast superimposed over the digitally enhanced photo of the bite mark. So, it's not as if prosecutors and defense lawyers won't try to get sketchy stuff admitted, it just a matter of when they do.

Yes (1, Informative)

Anonymous Coward | more than 3 years ago | (#35969626)

Yes [dartmouth.edu]

Re:Yes (1)

e9th (652576) | more than 3 years ago | (#35970138)

Thanks. There was only one useful item there, involving Tom Sizemore's conviction for beating up Heidi Fleiss. His attorneys claimed a photo of her injuries was faked. The judge gave prosecutors 60 days to either prove it was real or produce the photographer. Unfortunately, his conviction is still being appealed, and searching for further news takes me to more celebrity news sites than I can take.

What this makes me think is that if the police produce an image of you doing evil, all they need to do is have the photographer show up and swear that it hasn't been altered. This may be why Canon and Nikon don't seem too worried.

Re:Yes (1)

AndersOSU (873247) | more than 3 years ago | (#35970598)

and really, we're better off that way.

Photographic and video evidence are already incredibly compelling. We don't need some forensic expert telling the jury it can't be faked. We need the photographer telling the jury he didn't fake it.

Re:Yes (1)

LordLucless (582312) | more than 3 years ago | (#35972016)

There was also this:

August 2005: A magistrate in Sydney, Australia threw out a speeding case after the police said it had no evidence that an image from an automatic speed camera had not been doctored. This case revolved around the integrity of MD5, a digital signature algorithm, intended to prove that pictures have not been doctored after their recording. It is believed that this ruling may allow any driver caught by a speed camera to mount the same defense.

Re:Does it really matter? (2)

0123456 (636235) | more than 3 years ago | (#35969702)

Has there ever been a case whose outcome depended on the authenticity of a digital image?

If I remember correclty, three or four years ago a driver in the UK got out of a fine because he was able to prove that the photo used as evidence was faked. I don't remember the details, I think he parked in a car park and they tried to claim he overstayed using a doctored photograph as evidence?

Blog entry is down: database error, press release (1)

owlstead (636356) | more than 3 years ago | (#35969654)

Their press release can be found here:

http://www.elcomsoft.com/nikon.html [elcomsoft.com]

The press release does mention that you have to extract the key from the camera. If this is relatively easy then the system is totally broken. If it is not, you could create some kind of revocation list - but it would be the equivalent of a sloppy patch. Security is hard to accomplish, it does not surprise me that a camera manufacturer fails hardware protected signature creation.

Red light cameras? (0)

Anonymous Coward | more than 3 years ago | (#35969786)

Is this how those pictures are authenticated?

One Key? (1)

John.P.Jones (601028) | more than 3 years ago | (#35969906)

Disclaimer, I didn't RTFA.

Wouldn't each camera have its own signing key so all they could do is forge pictures from a single camera? They couldn't forge pictures from another camera without its key. Is there evidence of the key extraction left on the camera?

Re:One Key? (0)

Anonymous Coward | more than 3 years ago | (#35970026)

Can you imagine having to submit a camera as evidence that the pics are real in every case? You'd need a new one every couple days, I think effectively your distinction doesn't matter.

Re:One Key? (1)

Darinbob (1142669) | more than 3 years ago | (#35971056)

I haven't read the article, but much of this depends on who you're going to trust and who you suspect of tampering. Ie, if this is a surveillance camer in a bank, then you just have it securely mounted with tamper seals along with internal tamper protection; you trust the manufacturer to generate unique keys per camera when they sign a document claiming this is what they do; etc. This way when a defendant claims they weren't there and that the prosecutor doctored the photos, you have reasonable evidence to the contrary (and "reasonable doubt" is the standard used in courts, not "any extremely remote possibility"). So in essence you trust the bank manager and the camera manufacturer to not be in collusion with the police or DA.

Don't take my Kodachrome away (1)

NicknamesAreStupid (1040118) | more than 3 years ago | (#35970014)

Photographic film does not have this problem, though their prints can. Film contains a holographic image, albeit not like the dramatic ones you generally see. It has always been impossible to fake it.

Re:Don't take my Kodachrome away (0)

Anonymous Coward | more than 3 years ago | (#35970220)

http://en.wikipedia.org/wiki/Film_recorder

Take an original film image, scan it into a computer, manipulate it, then print it back onto the film with a film recorder. If you use a drum scanner and a high resolution film recorder, it would be extremely difficult, if not impossible, to tell the original image from the manipulated one.

Re:Don't take my Kodachrome away (2)

PopeRatzo (965947) | more than 3 years ago | (#35970398)

Film contains a holographic image

I didn't know that. Can you point me to any information about this? I'm googling here and not coming up with anything about emulsion negatives containing holograms, but probably because I'm not formulating the search very well.

If you have any links I'd really appreciate it.

Re:Don't take my Kodachrome away (2)

adolf (21054) | more than 3 years ago | (#35972946)

If you think of the emulsion layer as being a three-dimensional object that has some depth to it, instead of a two-dimensional plane with zero thickness, I believe that you'll find that it is obvious: It will not be exposed equally throughout that depth, and there will be definite and observable paths that the light has followed within the emulsion layer.

I don't know if I'd call it "holographic," just due to the confusion that the term itself presents in common use (as GP pointed out), but it seems like an adequate and correct description nonetheless.

Re:Don't take my Kodachrome away (1)

PopeRatzo (965947) | more than 3 years ago | (#35973154)

If you think of the emulsion layer as being a three-dimensional object that has some depth to it, instead of a two-dimensional plane with zero thickness

Thank you for giving me the opportunity to learn a little bit about what "holographic" means.

It's one of the countless areas in which I have been mostly ignorant. This is why I like reading Slashdot.

Re:Don't take my Kodachrome away (1)

adolf (21054) | more than 3 years ago | (#35974212)

Naah. Thanks, to you, for raising the question.

I hadn't ever had a reason to think about it in this context until you questioned the issue. And since you did, I got to learn something myself. It happened to be immediately obvious to me, in my little pea-sized brain, and it seems my brief description has made it obvious to you as well.

But again, I'd never have thought of it if you weren't curious yourself. We both learned a bit.

Re:Don't take my Kodachrome away (1)

The13thSin (1092867) | more than 3 years ago | (#35974962)

Interesting, though I'm somewhat skeptical at the idea that any film at any iso has the resolution to actually retain any usable information from that. Also consider that when something is shot with a small aperture, all the light comes from pretty much one point (think about how lenses work) and even with a large aperture, the difference in direction is not spectacular. This makes me further doubt that film (especially in 35mm) could have any further information (that is detectable with any normal forensic equipment) that would allow you to detect tempering.

Re:Don't take my Kodachrome away (2)

$RANDOMLUSER (804576) | more than 3 years ago | (#35970616)

"Kodachrome" is exactly right. I worked in a large (3M) film processing lab (factory, really) back in the early 70's. Probably once a week, the local sheriff or PD would have an officer come by with some SLIDE (chrome) FILM (typically autopsy or crime-scene photos) to process - they'd stand by and watch while their film was processed - maintaining the CHAIN OF CUSTODY [wikipedia.org] at all times, and requiring signatures from workers when the film was out of sight (like in a darkroom). They never had their slides mounted, they'd just walk away with the whole processed roll. I don't know for a fact, but I'd be willing to bet that it's still the standard way to handle film evidence (that really matters) even today - except on NCIS or CSI.

Re:Don't take my Kodachrome away (1)

LaminatorX (410794) | more than 3 years ago | (#35970692)

1) "Holographic" does not mean what you think it means.

2) There is a device called a film recorder which uses lasers to draw an image onto film from a digital file.
An expert might, be able to detect it examining the film under a microscope.

Re:Don't take my Kodachrome away (1)

MichaelSmith (789609) | more than 3 years ago | (#35971636)

1) "Holographic" does not mean what you think it means.

My reading of that post is that the track made through the depth of the film should tell you something about the distance to the object being photographed and the optics being used. Makes sense to me. Film is never 2D

Re:Don't take my Kodachrome away (1)

pz (113803) | more than 3 years ago | (#35971648)

Hmm. Interesting that you would say so. Positives were routinely retouched and then shot onto internegatives that were then re-shot onto a print. For a properly shot interneg, you cannot tell that it was not a shot of the original image (as opposed to a dupe).

so stupid (1)

holophrastic (221104) | more than 3 years ago | (#35970242)

and what stops me from taking a real photograph of a fake photograph? snap, photoshop, print, snap.

no one cares if the "photograph" is real or fake. We care if the content of that photograph is real or fake. So unless they digitally sign the universe to match the photograph, they've done nothing.

not to mention, have people forgotten that there are other ways to fake a photograph than with photoshop? Ever heard of actors, sets, studios, and lighting? Glass paintings, forced perspective, and dry ice?

again, congrats on trying to say that no one used photoshop since the last snap of the camera. That's really not difficult to keep true.

Re:so stupid (1)

owlstead (636356) | more than 3 years ago | (#35970430)

"and what stops me from taking a real photograph of a fake photograph? snap, photoshop, print, snap."

I'm pretty sure that you can't rule this out, but I also think that there are many ways of messing that up. Pixel artefacts, discolouring, stripes, loss of resolution, seems between separate parts of the picture etc. etc. may make life more difficult. And many printer/copiers even deliberately add water-signs or other identifying features (HP yellow dots for instance).

Re:so stupid (1)

holophrastic (221104) | more than 3 years ago | (#35970580)

ok, skip the print. photograph of a monitor. certainly it's not as easy as not having to do it; but we're not talking about someone's vacation photos. we're talking about something worth money, and hence time to forge.

Re:so stupid (0)

Anonymous Coward | more than 3 years ago | (#35970974)

I can't claim to know much of anything about optics, but I'd imagine that there's still likely to be some level of distortion due to the lens taking a picture of a two-dimensional image of a three-dimensional scene. If you're really good about it, it may not be perceptible to the naked eye, but I doubt it would stand up to scrutiny from an expert in the field.

An additional problem is that on most high end digital cameras (hell, most point-and-shoot cameras), the lens model, focal length, shutter speed, aperture setting, time, and all sorts of other meta-data is recorded with the picture. Even with a pretty nice studio, I'd figure it'd be pretty hard to take a picture of a printout/monitor image to replicate the desired image with the meta-data resembling anything that would be realistic for the image produced.

Re:so stupid (1)

holophrastic (221104) | more than 3 years ago | (#35971418)

you can delete the meta data entirely, or save the image in a format that doesn't store meta data at all.

As for the 2d ccd sensor taknig a 2d photograph of a 2d image, in theory it's got to be perfectable.

Re:so stupid (1)

starbuzz (590877) | more than 3 years ago | (#35972054)

Also, EXIF data [wikipedia.org] .

The image metadata should match all exposure settings suitable for the scene. This is difficult to achieve in reproduction.

Did these guys violate the DCMA in doing so? (0)

Anonymous Coward | more than 3 years ago | (#35970298)

Curious if these guys violated the DCMA in the course of doing this? Sounds kind of like the printer cartridge manufacturers that reverse engineer the code on printer cartridges intended by Canon, HP, et al to ensure that only OEM-branded cartridges can be used.

Re:Did these guys violate the DCMA in doing so? (1)

blueg3 (192743) | more than 3 years ago | (#35971148)

Elcomsoft? The Russian company Elcomsoft practically lives to violate the DMCA.

Re:Did these guys violate the DCMA in doing so? (0)

Anonymous Coward | more than 3 years ago | (#35971712)

Everybody forget about Dmitry Skylarov? That's the company he was an employee for at DefCon a few years back when he got busted for the Adobe DRM crack.

Re:Did these guys violate the DCMA in doing so? (1)

Sloppy (14984) | more than 3 years ago | (#35974970)

If you have the camera in question, then you can take photos for which you are the copyright holder. Then as the copyright holder, you can authorize yourself to bypass the technological measure that limits access (wait, this measure doesn't limit access) to your photo, which makes the activity not be "circumvention" (as DMCA defines that word).

So basically: no and no.

105mm f/2.5 (0)

Anonymous Coward | more than 3 years ago | (#35970574)

The 105mm f/2.5 causes me forgive any mistakes Nikon may have committed. And I want it that way.

Re:105mm f/2.5 (1)

webmistressrachel (903577) | more than 3 years ago | (#35970730)

I find that the 20+ year-old 70-210mm f/3.5 that I got with my 2nd-hand Nikon D70s purchase provides everything from a wide enough angle for landscapes to close zooms for shooting police activity. This and the 5 or so frames per second forces me to feel exactly the same way - I don't care about any of the DRM/JPG crap not working or being "forgable" as long as they don't ever, ever take away my RAW - which isn't editable anyway as it's a one-way format like a hash - raw data from the sensor, tagged with the current camera settings and adjustable losslessly in post.

In other words, This news isn't going to put me (or most other shooters) off Nikon.

Air Jordan 23 (-1)

Anonymous Coward | more than 3 years ago | (#35971032)

Air Jordan store has recently lanch online, providing high standard, high-class service and it also offers various kinds of products,such as Air Jordan 1,Air Jordan 3,Air Jordan13 which with its fashionable unique attracted numerous shoppers.The Cheap Air Jordan in order to satify different levels of demands it discount some of it's goods like Air Jordan Shoes.If you have desire don't miss occuring only once in a thousand years chance

fuck nikon (-1)

Anonymous Coward | more than 3 years ago | (#35971820)

Fuck is Nikon wasting developer resources on shit like then, when they could be doing something useful, like giving us 64 bit nef drivers.

Elcomsoft makes excessive demands (1)

Anonymous Coward | more than 3 years ago | (#35972570)

I think Elcomsoft is too strict and too harsh versus the digicam vendors. It is not possible to design a secure device at all, if the private key is stored in the device. An adversary advanced enough will have superb lab gear, including a scanning tunneling electron microscope and can sort through the integrated circuits atom by atom, if necessary. The key will be retrieved eventually.

Even if that does not work, the attacker could monitor power consumption or other side channel signals to deduce the keys. It is not reasonable to except a commercial digital camera to have TEMPEST features en par with a KH-11 spy satellite!

I think photo authenticity should be provided by special digicams, which also shoot a 135mm common film frame, in addition to the higher quality CCD/CMOS digital image. Chemical based roll film megative contains an almost infinite amount of information, so experts could you use it determine or exclude fakery.

Use smart cards! (1)

Anonymous Coward | more than 3 years ago | (#35973026)

This is what you get for implementing your own crypto. My suggestion to both Nikon and Canon is to include an ISO-7816 ID-000 port in their cameras (more commonly known as the SIM slot in mobile phones) and support one of the well-defined standards for public key operations on smartcards (PKCS#11 for example).

This means they have far less pressure to build a robust cryptographic system as it is built-in to the many, many compliant and certified smartcards out there in the market. Instead they can concentrate on the much simpler problem of using the crypto functions properly.

The major added benefit is that customers with strict security requirements (i.e. law enforcement) don't have to trust the crypto engines/key storage mechanisms developed by their camera supplier. They can put their own (trusted) smart-cards in the camera as simply as swapping a SIM in a mobile phone. For customers that don't want this, they can simply trust the default smartcards supplied with the camera.

Re:Use smart cards! (0)

Anonymous Coward | more than 3 years ago | (#35975682)

They're not interested in providing an strong, secure method of photo signing.

They're interested in taking an off the shelf camera they already make, bundling it with custom firmware, and selling for a massive markup to a specialty market with deep pockets.. Preferably with lucrative support contracts, attached software licensing, etc.

Load More Comments
Slashdot Login

Need an Account?

Forgot your password?