biometric

Biometric refers to the measurement and analysis of unique physical or behavioral traits used to identify individuals. In security and forensics, biometric systems compare a claimed identity with a stored template derived from a trait to grant or deny access. Methods rely on inherent characteristics such as fingerprints, iris patterns, facial features, voice, gait, or DNA, and often combine traits to improve accuracy.

Common modalities include fingerprint, iris, facial recognition, voice verification, and hand geometry. Each modality generates a digital template from a sample and uses algorithms to compare against stored templates. Verification confirms a claimed identity; identification searches for a person among many. Many systems store templates rather than raw data to reduce privacy risks and may perform matching on device or on a secure server. Liveness detection and anti-spoofing address fraud.

Biometric systems are used in mobile devices, workplace access, border control, and time and attendance, as well as in law enforcement and some commercial applications. They offer convenience and strong access control but raise privacy, consent, and data security concerns. Performance is described by false acceptance rate, false rejection rate, and equal error rate; accuracy varies by trait, population diversity, and environment. Bias and accessibility issues can arise if datasets are unrepresentative.

Regulation emphasizes data minimization, consent, retention limits, and security. Standards bodies work on interoperability and privacy-preserving techniques. Ongoing research seeks to improve accuracy, reduce bias, and enable privacy-preserving enrollment and matching, including on-device processing and template protection.