Tuesday, 30 July 2013

LIMITATIONS OF BIOMETRIC SYSTEM

                                                            In recent years, biometric systems have been productively deployed in a number of real-world applications with some biometrics contribution rationally good overall presentation. However, even the most highly developed biometric systems to date are opposite numerous problems, some inherent to the type of data and some of them natural to system design. In particular, biometric systems generally suffer in person verification process due to the factors listed below.
Sound can be defined as unnecessary data without meaning connected with the data. Noisy data is one of the common troubles of biometric systems. Noise can be incorporated in the biometric data during achievement due to defective, inappropriately maintained or outdated sensors, due to the malfunction of providing noise free biometric data achievement atmosphere or simply produced as an unnecessary by-product of other behavior. For example, capturing voice biometric data in noisy surroundings (i.e. during heavy rain, etc.) will result in a noisy voice signal enrolment. While an undesirable trend, the recognition accuracy of a biometric system could be responsive to the quality of the biometric data.
This chapter provides reader with a general idea of various biometric notions and terms, particularly physical and behavioral identifiers. Among the biometric identifiers, face, fingerprint, signature, voice and iris are the most used biometric identifiers due to the ease of their accessibility and their acknowledgment presentation. New categories of biometric identifiers, namely soft biometrics and social biometrics, are also introduced.

This chapter also discusses biometric functionalities and presentation parameters. Biometrics verification can be subdivided into authentication and recognition. Due to the nature of the application, the user or developer must decide on the suitable system architecture. The typical biometric system structural design is introduced next. Along with known modules, new approaches to intellectual decision-making, feature withdrawal, pattern identification, and system learning are outlined. The overall trend over recent few years was to recompense for inherent issues with biometric data and system presentation through introducing fundamentally new methods based on intellectual information fusion and intellectual pattern recognition.

No comments:

Post a Comment