PhoneAbility

Biometrics - introductory key concepts and review of aspects relevant to accessibility

Clive Reedman

Slide 2
Content

An introduction to iAFB
An introduction to Biometrics and their application
The implementation success factors
A future view
Questions

Slide 3
The iAfB (a few words)

The iAfB is the largest representative biometrics body in the world!
Current membership is now 90+ and rising
Membership spans industry, academia and government
- Our services to members:
- Trade Missions
- Support for the creation of a UK Centre of Excellence
- Focussed workshops
- Promotion of best practices, etc…

Slide 4
The iAfB (a few more words)

iAfB members are directly involved in:

Standards formulation
- ISO/IEC SC/37
- BSI IST/44
Test and evaluation
- Biometrics Working Group
Major government Identification projects
- UKPS/IND/IDENT1/PITO
- Research and development
- UK Centre of Excellence in Id Technologies

Slide 9
A Definition (One of Many)

Biometrics: The automated identification, or verification of human identity through repeatable measurements of physiological and behavioural characteristics

Slide 10
Identification v Verification

Identification:
A search of existing (enrolled) subjects:

Do I know you?
"I am not who I say I am not!"

Verification:
A 'one to one' comparison:

Are you who you say you are?
"I am who I say I am"

Slide 11
Some Examples

Physiological

Fingerprint
Face (2D/3D/5D)
Iris
Vein Pattern
Hand/Finger Geometry
etc

Behavioural

Signature
Gait
Speaker Verification
etc

Curently fingerprint, face and iris are the most commonly considered Biometrics.

Slide 12
Some Facts

There is no one 'killer' Biometrics
Biometrics are nearly always a part of a larger system
The 'something you are' is very personal to the individual
It is 'solution' and not 'technologies' are what should be promoted
There is no substitute for experience and good project management
People can be excluded from the 'norm'
People can be included where thay were previously excluded!

Slide 13
The Origin of Errors (FAR/FNMR)

False Accept Rate
(FAR)
The bad guy gets in!

False Non-Match Rate
(FNMR)
The good guy is locked out!

Slide 15
The Origin of Errors (FTE)

Failure to Enrol
The Biometrics cannot be captured, or captured well enough for the system to be able to use it.

Why?

There are many reasons, the majority related to the physiological (or other) characteristics of the user 'set'.

Slide 17
The First Principle!

GOOD PROJECT MANAGEMENT IS PARAMOUNT

When proposing a Biometric solution the issues are more often than not legal, political and social. It is therefore essential to understand and document the requirements and understand the implications before any Biometric technology solution is chosen (if at all)

Slide 18
User Attitude

Biometric systems may be thought of as a marriage between technology and human beings.
In any good relationship though, if one of the partners becomes neglected/undervalued, the relationship will suffer. You cannot underestimate the human aspect of the equation in your biometric application, and you must satisfactorily find the answer to this question:

How well do you know your user population?

Slide 19
Further 'Soft' Issues

User Perception
- Intrusiveness?
Privacy Concerns
- Loss of privacy/Big Brother?
Target Clientele Considerations
- Gender/age/disability etc.
User Difficulties
- Inherent characteristics of user population
Ease of Use
- Can (will) they use it?

Slide 20
Other Considerations (1)

Positive, or Negative Identification?

+ To prove I am someone registered on the system

- To prove I am not someone already registered in the system

+ Comparison of submitted sample to a single claimed template

+ /- Comparison of submitted sample to multiple, similar looking templates to look for a possible duplicate

Slide 21
Other Considerations (2)

Co-Operative v Non Co-Operative Users?

Is the user expected to co-operate (a good guy)?

Is the user not expected to co-operate (a bad guy)?

Does the user want to be identified, or not?

Does the user feel threatened?

Does the user feel embarrassed, or inconvenienced?

Slide 22
Other Considerations (3)

Habituated, or Non-habituated Users?

How often will users interact with the system?

Is use of the system intuitive?

Does the user have to remember how to use the system?

Is there a training issue?

Access control to a social service, once a month cannot be considered at 'habituated'.

Slide 23
Other Considerations (4)

Supervised, or Unsupervised Application?

Will somebody be there to help?

Are disabled users able to call on help?

Are operators trained to deal with special needs?

Slide 24
Other Considerations (5)

Standard, or Non-standard Environment?

Is the system used indoors, or outside?

The environment effects the users of their personal characteristics.

Is physical impairment catered for in the physical environment?

Slide 25
Other Considerations (6)

Overt, or Covert Use?

Does the user know that their Biometric is being captured?

Does a covert capture meet the legal requirements?

Slide 26
Where We are Heading

Government will provide the growth environment, but industry will be lead by popular demand.

Biometrics will become 'ordinary' for most users.

They may even become the next 'must have'.

Slide 27
The Future Vision (Ambient Biometrics)?

But it must be for all..This is the real challenge!

Contents page