i. Biometric authentication is one of the most exciting technical improvements of recent history and looks set to change the way in which the majority of individuals live.

ii. The use of biometric systems for personal authentication is a response to the rising issue of authentication and security. The most widely used method of biometric authentication is fingerprint recognition.

iii. Biometric authentication is considered the automatic identification or identity verification of an individual using either a biological feature possesses physiological characteristics like a signature.

iv. Biometric characteristics can be separated into two main categories:

• Physiological Characteristics: They are related to the shape of the body. The trait that has been used the longest, for over one hundred years, are fingerprints, other examples are face recognition, hand geometry and iris recognition.
• Behavioural Characteristics: They are related to the behaviour of a person. The first characteristics to be used that is still widely used today is the signature.

Figure 6. 4 Physical and Behavioural characteristics used by biometrics

v. A simple Biometric System consists of four basic components:

a. Sensor Module which acquires the biometric data.

b. Feature Extraction Module where the acquired data is processed to extract feature vectors.

c. Matching Module where feature vectors are compared against those in the template.

d. Decision-making Module in which the user’s identity is established or a channel identity is accepted or rejected.

vi. Any human physiological trait can serve as a biometric characteristic as long as it satisfies the following requirements:

a. Universality. Everyone should have it.

b. Distinctiveness. No two should be the same.

c. Permanence. It should be invariant over a given period of time.

d. Collectability

e. In real life applications, three additional factors should also be considered: performance (accuracy, speed, resource requirements), acceptability (it must be harmless to users), and circumvention (it should be robust enough to various fraudulent methods.)

Figure 6.5 Functional Schematic of Biometric Authentication System (BAS)

vii. Functional Model of Biometric Authentication System (BAS) :

As shown in figure 6.5 above:

• Biometric samples are collected using an appropriate sensor. The samples are then processed to correct the deterministic variations like translational and rotational shifts due to interaction of a sensor with the external world. This leads to set of “discriminatory” attributes that are invariant to irrelevant transformation of the input at the sensor.
• Following this segmentation/identification is performed to extract/recognize the desired attributes from the biometric samples.
• Measurements performed on these attributes give features depending upon the representation method.
• The features so obtained are used to form a biometric template. The biometric template is stored in one of the many encrypted forms so as to avoid spoofing.
• Once the database is ready, a query template needs to be authenticated using a matcher so as to determine its similarity with templates in the database.
• The output of the matcher is a matching score which gives the degree of similarity of the query template with various templates. This is used to arrive at a decision using a classifier.