Biometric Security

4th Year Computer Science

2022-2023
by
Prof.Dr. Ahmed AK Tahir

[email protected]
 Levels of Security
1- Possession Level
What You Have?
Token (Key, ID Card)

2- Knowledge Level
What You Know?
Password
PIN

3-Biometrics Level
Who You Are?
Biometric (Physiological, behavioral)
 Biometrics
The term "biometrics" is derived from the Greek words
bio (life) and metric (to measure).
It refers to the technology of measuring and analyzing a
person's physiological or behavioral characteristics,
which are called “human traits or human biometry”.
These human traits are unique to individuals hence can
be used to verify or identify a person.
 Biometric Types
1. Physiological Traits or Biometrics: are physical
measures or pattern on a person’s body.
 Iris
 Fingerprints
 Finger vein
 Face
 Hand vein

2. Behavioral Biometrics: are specific patterns that

are related to an individual’s actions.
Signature
 Gait
 Key stroke
 voice
Physiological Biometrics

Examples of biometric characteristics that are

commonly used: (a) face, (b) fingerprint, (c) hand
geometry, (d) iris.
Behavioral Biometrics

Examples of Behavioral biometric characteristics that are

commonly used: (a) Key Stroke, (b) Gait, (c) Signature, (d)
Vioce.
 Requirements to biometrics
1. Universality: everyone should have it

2. Uniqueness: small probability that two persons are the same in

terms of this characteristic

3. Permanence: invariance with the time

4. Collectability: can be measured quantitatively

5. Performance: high identification accuracy

6. Acceptability: acceptance by people “Non-intrusiveness: Imaging

without touch”

7. Circumvention: how easy to fool the system by fraudulent

technique

7
Comparison
Applications
 Terminologies
Traits: is a term used to refer to the biometric characteristic.

Enrollment: The initial process of collecting biometric data from a

user and then storing it in a template for later comparison.

False-acceptance rate (FAR): The percentage of imposters

incorrectly matched to a valid user’s biometric.

False-rejection rate (FRR):The percentage of incorrectly rejected

valid users.

Template: Is a term usually refers to a set of features derived from

the raw images of the traits.
 Biometric System
A biometric system is essentially a pattern recognition system
that operates by acquiring biometric data from an individual,
extracting a feature set from the acquired data, and comparing
this feature set against the template set in the database.
Depending on the application context, a biometric system may
operate either in verification mode or identification mode
 “Biometric Recognition System”
A typical biometric system is composed of five integrated
components.

Decision Matching
Enrolled Database

Authentication Enrolment
Mode Mode

Feature
Pre-Processing Pre-Processing
Extraction

Data Acquisition Data Acquisition

 Biometric System Stages

1. Data Acquisition: In this stage data is acquired via a sensor to be tested for
identification. The nature of this data depends on the biometric feature used
in the system. In iris and finger prints recognition system, this data is an
image of the iris or finger prints usually acquired by a camera device. In voice
recognition system, this data is a series of values representing the voice
signal. The quality of the acquired image or signal is important and crucial for
further identification/authentication task. In some biometric systems, namely
iris and face recognition systems, more than one image is required in order
to reach the best matching with the stored database.
 “Biometric Recognition System”
2. Feature Extraction: This stage covers the process of biometric templates
(traits) determination upon which the process of recognition can be
evaluated. Biometric template is usually extracted from the input data and
it should be very representative.
3. Storage: After feature extraction stage, the digital representations of the
templates are stored either in a central database or in a card for further
identification and authentication operations.
4. Matching: In this stage the comparisons are done between the stored
templates and the input template (the test template). The methods of the
matching stage should be capable of providing the correct identification.
5. Decision: In this stage the final decision about the involved case is taken
according to the output of the previous stage.
 Recognition Modes
The two main modes of biometric recognition system are:

 Verification
 Identification
 One to One comparison
Verification Mode
In the verification mode, the system validates a person’s identity Test User Enrolled Data
by comparing the captured biometric data with her own biometric D A
template(s) stored system database. In such a system, an B
C
individual who desires to be recognized claims an identity, usually
D
via a PIN (Personal Identification Number), a user name, a smart E
card, etc., and the system conducts a one-to-one comparison to F
determine whether the claim is true or not (e.g., “Does this G
biometric data belong to claimed person?”).
Identity verification is typically used for positive recognition,
where the aim is to prevent multiple people from using the same
identity.
 Identification Mode One to ALL comparisons

Test User Enrolled Data

D A

B
In the identification mode, the system recognizes an individual by
searching the templates of all the users in the database for a match. C
Therefore, the system conducts a one-to-many comparison to
establish an individual’s identity or fails if the subject is not enrolled D
in the system database. Identification is a critical component in
E
negative recognition applications where the system establishes
whether the person is who she (implicitly or explicitly) denies to be F
in the black list. When Identification is used for negative recognition
it is called as screening. The purpose of negative recognition is to G
prevent a single person from using multiple identities.
 System Evaluation
The performance (Accuracy) of the system is evaluated from the following
measures:

True Acceptance Rate (TAR): Proportion of genuine users accepted

 False Accept Rate (FAR): Proportion of imposter users accepted
True Rejection Rate (TRR): Proportion of imposter users rejected
 False Rejection Rate (FRR): Proportion of genuine users rejected
 Biometric Data Acquisition Devices
Human biometric trait data is usually acquired via a sensor, camera or recording
devices depending on the nature of the trait. For iris and fingerprints, the data
is in image form acquired by Infrared camera devices, whereas for voice data is
a series of values representing the voice signal taken by voice recorders. The
quality of the acquired image or signal is important for accurate
identification/verification. To increase the performance of biometric systems,
more than one image of the human trait is required to reach the best matching
with the enrolled database. The quality of the data depends on the sensor
specifications and the imaging conditions. For example, the optical fingerprint
imaging, which captures a digital image of the fingerprints using visible light is
affected by the quality of skin on the finger, while Ultrasonic sensors use very
high frequency sound waves to penetrate the epidermal layer of skin. Also,
thermal scanners can be used to sense the temperature differences between
fingerprint ridges and valleys, so it is less affected by the skin quality.
The table below shows some of the data acquisition device and some of the
standard datasets which can be used for research development.
 Biometric Acquisition Devices
Biometric Acquisition Devices Available Database for research
traits development
Iris Infrared Camera CASIA Version 1-4, SDUMLA, UBIRIS
Finger Infrared Sensor SDUMLA, HKPU,
Vein Near-infrared LED light UTFVP, MMCBNU_6000
Monochrome CCD
camera
Fingerprin Sensors: CASIA, SDUMLA, Multi-Sensor Fingerprint
ts Optical camera Database
Ultrasound scanner
Thermal scanners

Face Camera SDUMLA-Face database part 1 and part 2.

Voice Voice Recorder + A/D VoxCeleb, 2000 Hub5, Google Audiset and
Convertor TIMIT database
Signature Scanner digitizer SVC and MCYT, CEDAR, MCYT-75, GPDS and
UTSIG.