Creating and sharing knowledge for telecommunications

Project: NECOVID: Negative Covert Biometric Identification

Acronym: NECOVID
Main Objective:
The main purpose of this project is to address the feasibility of an automated system that performs covert and reliable biometric recognition, using the iris as single trait. Obviously, this type of biometric recognition is extremely ambitious and brings many challenges to the pattern recognition task, namely due to the many types of non-deal images that result of the imaging conditions and acquisition protocols (at-a-distance, on-the-move and under dynamic lighting conditions)
The proposed approach to deal with these extremely challenging conditions is based on the concept of negative (a-contrario) recognition, i.e., to prove that an individual is not among a group of people already known to the system. The key insight is that although the quality of the captured data possibly denies the positive recognition with enough confidence, perhaps it is still possible to assure that data is not correspondent to a subset of the enrolled templates, which for most of the everyday situations is the essential. The range of potential applications to the type of system proposed for this project is obvious. Indeed, this type of applications is regarded for the Pattern Recognition community as ‘’the grand-challenge’’ (e.g., ''Biometrics: a grand-challenge'', A.K. Jain), due to the implications that they can have in modern societies.
Reference: PTDC/EIA-EIA/103945/200
Funding: FCT/PTDC
Start Date: 01-11-2009
End Date: 01-11-2012
Team: Hugo Pedro Martins Carriço Proença, Luís Filipe Barbosa de Almeida Alexandre
Groups: Pattern and Image Analysis – Cv
Local Coordinator: Hugo Pedro Martins Carriço Proença
Internal Page

Associated Publications