Biometric recognition: Sensor characteristics and image quality

Biometric recognition, or simply biometrics, refers to recognizing a person based on one or more of his anatomical or behavioral characteristics. A good biometric trait should be measurable, distinctive (different for every person) and stable over time. The sensing method should not be intrusive or socially unacceptable, and the system should be easy to use. A biometric system based on this trait should be accurate, fast, robust and inexpensive. In this article, we discuss the signal acquisition aspects of fingerprint and iris biometrics-two of the most widely used biometric traits.Personal recognition of people is necessary to conduct many social and economic activities. Besides visual recognition of acquaintances, checking a person's government issued photo ID is the most common procedure. In electronic access or transactions, passwords and security tokens are commonly used. These credentials are surrogates of a person's identity Their major shortcoming is that they can be easily compromised by being lost, stolen, or given to someone else. For better security, it is necessary to link the digital identity of a person to his body's characteristics.