Security for Dummies

security tips and tricks

Faking a fingerprint (Part 2)

In my previous post I have described the simple way of faking a finger for an optical fingerprint sensor.
The second experiment demonstrated the high identification capabilities of optical fingerprint capturing devices together with their absolute lack of discrimination of fake “fingers”.

During the study we managed to get the fingerprint image that was accepted by an optical sensor. The “victim” was unaware and the fingerprint was captured during his absence.
The equipment for the second study was as simple and available as the previous one. After the clean and smooth surface contact the fingers of a “victim”, fingerprints are settled on it.

The only problem is to make them visible.
We used the glass (touched by the “victim”), scotch tape, PPC film (the same one as in the previous study), and carbon powder. You can use black toner for any laser printer or copy machine toner.

Gently sprinkle the surface with the carbon powder. Blow away excess powder. Aren’t they beautiful? You need to take them off the surface. Of course, we have scotch tape for this purpose.

Place the scotch tape above the fingerprint with its adhesive side towards the glass and gently stick the tape. Do not wipe or press it.
Gently pull the tape off. The fingerprint stays on it. Stick the tape to the PPC film. Repeat the procedure for each fingerprint. 
 
Now you have at least three fingerprint images. Cut them as in the previous experiment. You can run the identification procedure. They will undoubtedly be accepted.
 
The fingerprints of an innocent “victim” were captured and saved. They were identified later with an optical sensor and accepted without any hesitation.
 

CONCLUSION
Optical sensors are the most widely used devices in the field of fingerprint identification and the only type that has already found its consumer use. This makes them a target for falsification attempts. Our experiment demonstrated that optical sensors are not able to determine whether the presented for identification is a finger or an image. This makes it possible to reproduce fingerprint images using even a consumer copier. This is compromising the entire method of personal identification by fingerprint – making it untrustworthy and less reliable. The suggested alternative for optical biometrics would be to solve this problem by abandoning the optical sensor and turning to more reliable capacitive or thermal type of sensor.
Please notice that this was a conclusion that we came to a decade ago. It was partly correct. Today we know much more about biometrics. There are more sensors available. New optical sensors have an option of latent finger removal and a protection against artificial images. RF sensors like these has 47 patents for liveness detection. Manufacturers of capacitive sensors claim that only live fingerprints can be scanned. All this is bulshit. We claim that we can fake every sensor. Even now.
Whar is the bottom line? If you are concerned about your data never trust biometrics. Use it in combination with encryption, password and hardware factor. In later posts I will show you more tricks and more sensors. Keep reading.

 

Advertisements

June 27, 2008 - Posted by | Biometrics, How-to, Security | ,

6 Comments »

  1. Thanks to Karen Friar for the link to Dialogue Box video where Rupert and Charles make a fake finger out of plastic, jelly, milk and tea and try it out on a biometric security device? It’s a good laugh

    Comment by patholog | June 30, 2008 | Reply

  2. wow thats amazing that you can get away with all these things, and sad for those companies that falsely claim that its fool proof practically.

    Comment by Jaret | June 30, 2008 | Reply

  3. Yes, Jaret, you are right. But my aim is not to compromise biometrics. I just want to position it correctly. Every user shall understand that biometrics provides user convenience and ease of operation. Biometrics has nothing to do with security. It is a part of different science. It can be applied to security, but as add-on, and in no way as a main tool.

    Comment by patholog | June 30, 2008 | Reply

  4. Patholog,

    Can you post further information on the “foolability” or failure rate of the more advanced fingerprint scanners that you describe at the end of this post? I don’t want to fool them, I just want to know more about these devices which are being presented to the public as so reliable that we should accept their widespread use. My nightmare is that daily life will become similar to security at the airport: a big delay and irritation, which does not really provide much security.

    Comment by nosleepingdog | July 2, 2008 | Reply

  5. I can try to rate them from 0 to 10, where 0 – take seconds; and 10 – take hours or even days to fool a sensor, according to my experience
    I will not mention manufacturer and sensor name, not to cause envy of any of them, – only types of sensors
    Matrix optical – 0
    Matrix optical reflective -0.3
    Swipe optical -0.3
    Matrix capacitive – 2 to 6
    Pressure – 3
    Optical with MRI and blood flow control – 3
    Thermal -2
    RF matrix – 6
    RF swipe – 7

    It is likely that other tester will grade them in another way.

    Comment by patholog | July 2, 2008 | Reply

  6. This design is wicked! You obviously know how to keep a
    reader entertained. Between your wit and
    your videos, I was almost moved to start my own blog (well, almost.
    ..HaHa!) Great job. I really enjoyed what you had to say, and more than that, how you presented it.
    Too cool!

    Comment by megagamers.sosblogs.com | April 27, 2013 | Reply


Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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 )

Google+ photo

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

Connecting to %s

%d bloggers like this: