Why blurring sensitive information is a bad idea (2007) • Hacker News Board

Re: Why blurring sensitive information is a bad idea (2007)

#111

Post by pavel_lishin » Fri, Jul 25, 2014, 3:12 AM UTC

izzydata wrote: post #64

frogpelt wrote: post #54

Earlier quoted context omitted.

"just a pedophile" A phrase I never thought I'd see.

Someone who is homosexual isn't any more inclined to molest or rape someone of their own sex than a pedophile is to molest or rape a child. More on topic about censoring important information. I'm not entirely sure about this, but I feel like I've seen images where the thumbnail was retained from a previous version. Even at 32x32 there might be some way to expand it and gather some kind of information from its pixela…

We were using ImageMagick to resize images down to a smaller size. Turns out that, unless passed certain flags, ImageMagick would retain all metadata.

A few times, we'd end up with a 32x32 pixel image that was 20 megabytes in size - because that metadata included the original image.

Re: Why blurring sensitive information is a bad idea (2007)

#112

Post by varkson » Fri, Jul 25, 2014, 4:22 AM UTC

exelius wrote: post #57

jebus989 wrote: post #37

Earlier quoted context omitted.

Police == 4chan?

Yeah, basically. 4chan's admins have been working with the police for many years because of the questionable shit that gets posted on /b/ (child porn, murder threats/evidence, suicidal posts, etc.) They would have been shut down years ago if they weren't cooperating.

Going on around '09 it was pretty common to have pornography posted with the same type of swirl or other manipulations. It was a game to undo it, using a particular piece of software.

Re: Why blurring sensitive information is a bad idea (2007)

#113

Post by dheera » Fri, Jul 25, 2014, 4:27 AM UTC

shinigami wrote: post #36

nnnnni wrote: post #10

Their answer is to "color over it", but BE CAREFUL if you do that. I'm sure that some of us remember the US government document(s) in the late 90s/early 2000s that were released in redacted form, but the person didn't realize/understand that people with the Acrobat editor program could remove the black bars.

Also, image formats like JPEG have a thumbnail stored in them, which may not be updated when you edit the image.

Also, when blacking out data from JPEGs, leave generous margins, because there may be compression artifacts in the nearby pixels that also act effectively as a one-way hash. I haven't tested how consistent these artifacts are, but it's better to be safe than sorry.

Re: Why blurring sensitive information is a bad idea (2007)

#114

Post by sitkack » Fri, Jul 25, 2014, 4:29 AM UTC

nullc wrote: post #20

InclinedPlane wrote: post #19

Earlier quoted context omitted.

Are you a professional cryptographer? Can you prove that your random shuffling is impossible to extract sensitive data from? No? Then you better stick with completely deleting the data from the image entirely.

Brings back nausea from arguments past, that went something like this (dramatized for your pleasure): Me: Your protocol has a serious size-side channel that leaks all the important data as sizes. Please use a constant length encoding. Duh: Thanks! I added RANDOM padding. Totally secure now(tm). Me: Your random was rand(), I recovered the LCG state from a few packets with known sizes and then recovered the original si…

Don't forget that Duh also lights up the twitter and the hubs with their SuperSecretCoderRing. On the internet anyone can pretend to do brain surgery.

Re: Why blurring sensitive information is a bad idea (2007)

#115

Post by sitkack » Fri, Jul 25, 2014, 4:31 AM UTC

Udo wrote: post #60

jarman wrote: post #11

Earlier quoted context omitted.

It will be worse than that. Pixellation destroys information, it's not reversible in general. But in this case, what is left is enough for recovery, as source domain (check number) is small. Gaussian blur is theoretically lossless (rounding loss and loss on image edges in practice) - you can take arbitrary image, blur it, reverse and get original+noise

Thanks for making me look deeper into what's going on with common blurring algorithms. My instincts were clearly wrong here, for years I thought reversing blurs wasn't feasible in practice. I was very, very wrong.

Unknown Partial Knowledge is a huge problem in human cognition.

This blew my mind, https://www.youtube.com/watch?v=tyHBnT4PmTE

Re: Why blurring sensitive information is a bad idea (2007)

#116

Post by dheera » Fri, Jul 25, 2014, 5:46 AM UTC

robin_reala wrote: post #40

shinigami wrote: post #36

Earlier quoted context omitted.

Also, image formats like JPEG have a thumbnail stored in them, which may not be updated when you edit the image.

Do they? First I’ve heard of that…

You can use 'exiftool' on Linux to manipulate the EXIF data, get rid of the thumbnail, or set it to whatever you want. Imagemagick's 'convert' command also has a '-strip' option that will get rid of all EXIF data including any thumbnails.

Re: Why blurring sensitive information is a bad idea (2007)

#117

Post by dheera » Fri, Jul 25, 2014, 5:51 AM UTC

MichaelApproved wrote: post #31

nnnnni wrote: post #10

Their answer is to "color over it", but BE CAREFUL if you do that. I'm sure that some of us remember the US government document(s) in the late 90s/early 2000s that were released in redacted form, but the person didn't realize/understand that people with the Acrobat editor program could remove the black bars.

You don't have to go all the way back to the early 2000s to find poorly redacted federal documents. Rachel Maddow had a great segment on botched redactions, like removing the black lines from a PDF. The best solution was to actually cut out the parts of the page you don't want seen and then scan in the result. All other methods are prone to mistakes. Edit: corrected link to segment http://www.msnbc.com/rachel-maddow-…

Along the same lines, I remember an MIT Mystery Hunt puzzle once where there were hidden clues in a PostScript document. If you didn't think to look at the PostScript source you probably had no hope of solving it.

Re: Why blurring sensitive information is a bad idea (2007)

#118

Post by dheera » Fri, Jul 25, 2014, 5:53 AM UTC

im3w1l wrote: post #69

freehunter wrote: post #55

Earlier quoted context omitted.

I've always drawn over sensitive parts in Paint or something, then taken a screenshot of it. It makes sure that everything is compressed to a single layer and can't be undone.

Just be sure to get all of it. It's pretty common that people leave a few pixels. How not to do it: https://imgur.com/ZaCIY5P

Leave generous margins. JPEGs can have compression artifacts that leak information outside the boundaries of the object.

Re: Why blurring sensitive information is a bad idea (2007)

#119

Post by Natsu » Fri, Jul 25, 2014, 6:28 AM UTC

robin_reala wrote: post #40

shinigami wrote: post #36

Earlier quoted context omitted.

Also, image formats like JPEG have a thumbnail stored in them, which may not be updated when you edit the image.

Do they? First I’ve heard of that…

I believe it's a Photoshop thing, rather than something common to all images. Photoshop in general does unusual things.

Re: Why blurring sensitive information is a bad idea (2007)

#120

Post by Natsu » Fri, Jul 25, 2014, 6:32 AM UTC

seanflyon wrote: post #85

bithush wrote: post #23

Earlier quoted context omitted.

Indeed. See http://en.wikipedia.org/wiki/Christopher_Paul_Neil

Swirl is not a lossy process. If he had blurred his face he would have been much harder to catch.

It's partially lossy, from the look of it (you can see circular artifacts), but you're right that enough is recoverable to make it meaningless in terms of hiding data.