What are Biometrics?

Biometrics is a way to measure a person physical characteristics to uniquely verify or authenticate their identity. This can include physiological traits or behavior characteristics.

---

Why use Biometrics?

Population and mobility in modern world has made biometrics a desirable tool. For biometrics to be considered useful, data that is collected must be unique, permanent and collectable. Once measured, the information is compared and matched in a database.

---

Biometric Systems

A biometric system measures one or more physical or behavioral characteristics. These include physical features such as a fingerprint, face, iris or retina. As well as this, the voice and gait of a person can also be used for biometrical applications which can be used to determine or verify the identity of a person.

---

Generic Biometric System

Within a generic biometric system there are typically two main phases. The Enrollment Phase is the biometric data that is acquired from the individual and stored in database along with the persons identity. This is typically the data which is processed to extract relevant, salient distinctive features. The second of these steps is the Recognition Phase, this is the biometric data that is re-acquired from the user and compared against the stored data to determine the user identity.

A biometric system is essentially a pattern recognition or patter matching system consisting of four building blocks:

• Sensor
• Feature Extractor
• Database
• Matcher

---

Biometric Systems

An ideal biometric system must exhibit small inter-user similarity, and small intra-user variations. It must contain a threshold, which is the quality value that a subject’s features must meet before being considered as a successful authentication.

Biometric systems can make several types of errors. Three of the most important are:

• Failure to Enroll (FTE)
  • (Failed Enrollment Attempts) * 100/All Genuine Menthol Attempts
• False Rejection (FRR)
  • (Genuine Scores below Threshold) * 100/All Genuine Scores
• False Acceptance (FAR)
  • (All imposter scores above the threshold) * 100/All Imposter Scores

---

Performance Metrics

Failure to Enroll Rate (FTE): The rate at which attempts to create a template from an input is unsuccessful. This is most commonly caused by low quality inputs.

Failure to Capture Rate (FTC): The probability that the system fails to detect a biometric input when presented correctly.

Template Capacity: The maximum number of sets of data which can be stored in the database.

---

Crossover Error Rate

The Crossover Error Rate (CER) describes the point where the false reject rate (FRR) and false accept rate (FAR) are equal. This occurrence is also known as the Error Rate (ERR). However, CER is the optimal setting so that there is no disproportionate FRR or FAR.

---

Characteristics of All Biometrics:

1. Universality – Each person should have this characteristic. Face recognition has high universality.
2. Distinctiveness – Any two people should be distinguishable by this characteristic. Iris scanning has high distinctiveness.
3. Permanence – The characteristic should not vary over a period of time. Iris has high permanence.
4. Collectability – The characteristic can be easily measured quantitatively. Hand geometry has high collectability.
5. Performance – Recognition accuracy, speed, resources required, robustness to factors in a practical environment. Iris recognition has high performance.
6. Acceptability – Public acceptance of the technology in a given situation. Voice recognition has high acceptability.
7. Circumvention – How easy the system can be fooled. Retinal scanning is very difficult to circumvent.