The modern information age gives rise to various challenges, such as organization of society and its security. In the context of organization of society, security has become an important challenge. Because of the increased importance of security and organization, identification and authentication methods have developed into a key technology in various areas, such as entrance control in buildings, access control for automatic teller machines, or in the prominent field of criminal investigation. Identity verification techniques such as keys, cards, passwords, and PIN are widely used security applications. However, passwords or keys may often be forgotten, disclosed, changed, or stolen. Biometrics is an identity verification technique which is being used nowadays and is more reliable, compared to traditional techniques. Biometrics means “life measurement,” but here, the term is associated with the unique characteristics of an individual. Biometrics is thus defined as the “automated methods of identifying or authenticating the identity of a living person, based on physiological or behavioral characteristics.” Physiological characteristics include features such as face, fingerprint, and iris. Behavioral characteristics include signature, gait, and voice. This method of identity verification is preferred over traditional passwords and PIN-based methods for various reasons, such as (Jain, Bolle, & Pankanti, 1999; Jain, Ross, & Prabhakar, 2004): • The person to be identified is required to be physically present for the identity verification. • Identification based on biometric techniques obviates the need to remember a password or carry a token. • It cannot be misplaced or forgotten. Biometrics is essentially a multi-disciplinary area of research, which includes fields like pattern recognition image processing, computer vision, soft computing, and artificial intelligence. For example, face image is captured by a digital camera, which is preprocessed using image enhancement algorithms, and then facial information is extracted and matched. During this process, image processing techniques are used to enhance the face image and pattern recognition, and soft computing techniques are used to extract and match facial features. A biometric system can be either an identification system or a verification (authentication) system, depending on the application. Identification and verification are defined as (Jain et al., 1999, 2004; Ross, Nandakumar, & Jain, 2006): • Identification–One to Many: Identification involves determining a person’s identity by searching through the database for a match. For example, identification is performed in a watch list to find if the query image matches with any of the images in the watch list. • Verification–One to One: Verification involves determining if the identity which the person is claiming is correct or not. Examples of verification include access to an ATM, it can be obtained by matching the features of the individual with the features of the claimed identity in the database. It is not required to perform match with complete database. In this article, we present an overview of the biometric systems and different types of biometric modalities. The next section describes various components of biometric systems, and the third section briefly describes the characteristics of biometric systems. The fourth section provides an overview of different unimodal and multimodal biometric systems. In the fifth section, we have discussed different measures used to evaluate the performance of biometric systems. Finally, we discuss research issues and future directions of biometrics in the last section.
Incorporating Metadata in Multibiometric Score-Level Fusion: an Optimized Architecture