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defective, inappropriately maintained or outdated sensors, due to the malfunction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