|Research in Biometrics|
|Contact: Josef Bigun, Fernando Alonso-Fernandez|
Every human being has experience in recognizing a familiar person based on his/her specific characteristics, like voice, face, gait, handwriting, signature and so on. Some people, more than others, have even the ability to recognize unknown persons, after having seen or heard them. In the era of networked society, there as increasing need for reliable personal identification by automatic means. Establishing the identity of individuals is recognized as fundamental not only in numerous governmental, legal or forensic operations, but also in a large number of civilian applications. This has resulted in the establishment of a new research and technology area known as biometric recognition, or simply biometrics.
The term “biometrics” refers to automatic recognition of an individual based on behavioral and/or anatomical characteristics (e.g., fingerprints, face, iris, voice, signature, etc.). A biometric system essentially makes use of behavioral or anatomical characteristics to recognize individuals by means of pattern recognition techniques and statistical methods. Biometric systems offer greater convenience and several advantages over traditional security methods based on something that you know (normally a secret password or PIN, which can be shared, forgotten or copied) or something that you have (a physical object that is presented to receive access, such as keys, magnetic cards, identity documents, etc., which can be shared, stolen, copied or lost). Without sophisticated means, biometrics are difficult to share, steal or forge and cannot be forgotten or lost. Therefore, this latter solution provides a higher security level in identity prove. In addition, the combination of possession and knowledge with biometrics makes the identity proof even more secure.
- BBfor2: Bayesian Biometrics for Forensics
- BIO-DISTANCE: Biometrics at a Distance
- COST Action IC1106: Integrating Biometrics and Forensics for the Digital Age
- Facial Detection and Recognition Resilient to Physical Image Deformations