Holographic ID Recognition System

[petrushka.googol]

I was wondering if a 3D wire-frame of a person could be used to create a unique ID that could be detected by a holographic scanner and used as a valid proof of identity.

 

This 3D scan could be encrypted with a voice print key (that uses the frequencies most commonly occurring in any individuals speech) to produce a foolproof marker for an individual.

 

The voice print serves as the decryption key.

 

The holographic scanner will re-construct the 3D map of an individual and the concerned person will be asked to speak into a microphone that will sample the individuals speech on the algorithm discussed earlier.

 

This overcomes the possibility of phony ID's and is a "as-is-where-is" type of system. (very good at airports etc).

 

How feasible is this?

 

Please advise.

 

Thanks in advance.

 

[John Cuthber]

So, it won't work if I have a cold or sore throat which changes my voice and it will require me to undress a lot because my clothes might be bulky enough to mislead it.

 

Congratulations; you just made fingerprinting everybody at the airport look like a practical suggestion.

[petrushka.googol]

So, it won't work if I have a cold or sore throat which changes my voice and it will require me to undress a lot because my clothes might be bulky enough to mislead it.

 

Congratulations; you just made fingerprinting everybody at the airport look like a practical suggestion.

 

By using a sufficiently small sampling frequency we can eliminate local "noise".

Also, by considering the deviation from the mean produced by a sore throat, suitable correction can be applied to the waveform to arrive at a resultant waveform which can be reverse engineered to get the encryption key.

This could be achieved by using an oscilloscope plugged into a laptop. Hardware could be inexpensive and portable.

May be this is feasible?

[Stetson]

Facial recognition seems more feasible, less things to distort the face. As well there are so many unique features on a face, i.e. a scar, distance between key facial features. Or to get a little creative, hit them with x-rays and compare dental records on a database.

[imatfaal]

Again; back to John Cuthber's suggestion - what do benefits any of these ideas have over finger prints? Sure a few people lose them due to long term use of some prescription drugs or due to a terrible accident and a tiny number of people might be born without hands etc - but other than that everyone has them, they are all different, and they can be easily read.

[studiot]

Here is a learned paper on the subject

 

http://www.cs.cornell.edu/~dph/papers/HKR-TPAMI-93.pdf

[John Cuthber]

 

By using a sufficiently small sampling frequency we can eliminate local "noise".

Also, by considering the deviation from the mean produced by a sore throat, suitable correction can be applied to the waveform to arrive at a resultant waveform which can be reverse engineered to get the encryption key.

This could be achieved by using an oscilloscope plugged into a laptop. Hardware could be inexpensive and portable.

May be this is feasible?

It's not a matter of eliminating noise.

The original signal simply isn't there.

If you set the filters so broad that they will recognise my voice, even when I sound like Lee Marvin, then they will also recognise Le marvin as being me.

There's no discrimination left.

 

And, when you say things like "This could be achieved by using an oscilloscope plugged into a laptop." it tells me that you don't really know what you are talking about.

[petrushka.googol]

It's not a matter of eliminating noise.

The original signal simply isn't there.

If you set the filters so broad that they will recognise my voice, even when I sound like Lee Marvin, then they will also recognise Le marvin as being me.

There's no discrimination left.

 

And, when you say things like "This could be achieved by using an oscilloscope plugged into a laptop." it tells me that you don't really know what you are talking about.

 

Please refer to the following extract :

 

Voiceprint Identification, method used in biometrics by which a person purportedly can be identified from a spectrographic picture of his or her voice pattern when speaking words. In the same way that identification by fingerprinting assumes that no two persons have identical ridges on their fingertips, voiceprint identification assumes that certain physical characteristics of the vocal organs, which influence the quality of the sound in speech, will not be exactly the same in any two persons. Those characteristics are the size of the vocal cavities (the throat, nose, and mouth) and use patterns for the muscles of articulation in the tongue, jaws, lips, and soft palate.

 

Fingerprint ids are not foolproof. They can be spoofed to produce "fake fingers". My method is much more secure in theory.

[John Cuthber]

Please read the following extract of that extract

"method used in biometrics by which a person purportedly can be identified"

 

"My method is much more secure in theory."

Please read the definition of theory

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

 

 

Please read this account of how to use a laptop as a 'scope.

http://www.instructables.com/id/Use-Your-Laptop-as-Oscilloscope/

[imatfaal]

Please refer to the following extract :

 

Voiceprint Identification, method used in biometrics by which a person purportedly can be identified from a spectrographic picture of his or her voice pattern when speaking words. In the same way that identification by fingerprinting assumes that no two persons have identical ridges on their fingertips, voiceprint identification assumes that certain physical characteristics of the vocal organs, which influence the quality of the sound in speech, will not be exactly the same in any two persons. Those characteristics are the size of the vocal cavities (the throat, nose, and mouth) and use patterns for the muscles of articulation in the tongue, jaws, lips, and soft palate.

 

Fingerprint ids are not foolproof. They can be spoofed to produce "fake fingers". My method is much more secure in theory.

Ok - we will have a race to see who can make a spoof first, and which is most likely to succeed. Let's choose David Cameron the UK prime-minister as our target. Took me about 15 seconds to get an MP3 of David Cameron's voice. OK your move...