PRNU: How to Identify Your Smartphone from a Single Photo

Let’s talk about forensics, cameras, and how point & shoot cameras may actually protect your privacy a bit better than smartphone camera apps. Does that sound weird? It is – and we’re just getting started.

First, the forensics: it’s a process called PRNU, or photo-response non-uniformity. Basically, when you look at a photo, even a digital photo that’s never seen a print-out, there are tiny flaws all over it. Those flaws are created by many factors, but a number of them involve the software and hardware of the camera used to take the pictures.

You can probably see where this is going. Like identifying a fingerprint or a fired bullet, forensic specialists can use a photograph to identify the exact device it was taken on. They carefully examine all the pixels that are slightly different (usually lighter or darker than expected), and compare those pixels to new photos taken by devices to find a match. The differences are so small that humans can see them by looking, so specialized software does the job instead.

This can quickly identify what device took incriminating photos, for example – but here’s where it gets interesting. According to research done on both iPhones and Samsung Galaxy Note device, it’s far, far easier to identify smartphones with this process. For a regularly camera, it takes about 50 photos per camera to analyze before an authentic match can be made. For smartphones, however, you only need one photo and it can be successfully matched with the phone that took it. Results in either case are around 99.5% accurate.

This PRNU approach has potential for future ID protection too, so it’s not all about catching bad guys or going on a hunt for the right device. In fact, this could make it much more difficult to steal digital identities and use funds to buy stuff…if you have the wrong smartphone. Here’s hoping the tech will be get there quickly!

PRNU: How to Identify Your Smartphone from a Single Photo Find more on: GadgetReview (US)

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s