From time immemorial, the security realm and personal identification task had
shown progress by employing technological means like secret knowledge as passwords
and Personal Identification Numbers, and by using personal possessions as
identity cards and Radio Frequency Identification chips. As opposed to these means
which are generally easy targets for circulation and fraud, biometric traits (or modalities)
like facial geometry, iris, voice timbre, and biological DNA are universal,
difficult to copy, and for most, consistent over time with no expiration date and
comfortable to use. The purpose of this book is to provide an up-to-date ample coverage
of theoretical and experimental state-of-the-art work as well as new trends and
directions in the biometrics field.
The biometrics field – the science of measuring physical properties of human
beings – has marked a substantial leap over the past two decades. It will continue
to climb, as a result of a strong demand from both the private sector and governmental
agencies, on an ever ramping curve until practical objectives are achieved
in terms of high accuracy, ease of use, and low-cost. As the cost of biometric sensors
– visible, multispectral and thermal imagers, microphones, capacitive, pressure
and motion sensors – continues to sink due to higher demand, biometric systems
have the tendency to employ more than a single sensor to capture and identify an
individual upon as many as nonredundant biometric traits.
While this book covers a range of biometric traits including facial geometry, 3D
ear form, fingerprints, vein structure, voice, and gait, its main emphasis is placed
on multisensory and multimodal face biometrics algorithms and systems. By “multisensory”
we refer to combining data from two or more biometric sensors, such as
synchronized reflectance-based and temperature-based face images. Likewise, by
“multimodal” biometrics, we refer to fusing two or more biometric modalities, like
face images and voice timber. These two multisensory and multimodal aspects, as
they pertain to face Biometrics, are covered in details in the course of this book. Reported
experimental results support the idea that adequate fusion of complementary
biometric data is a step in the right direction to remedy to the limitations of existing
uni-sensor and uni-modal biometric systems.