A multimodal biometric system are developed for human identification. It is a project sponsored by the Ministry of Communication & Information Technology, New Delhi. The main objective of this project is to design a system that is capable of handling problems like security, personal verification and other related applications on the basis of biometrics features like face, fingerprint, voice and signature etc. The outcome is to verify or identity the person and if the person is not in the database then information is to be stored in the database for the future use. The system can be operated in two modes i.e.., verification and identification.
In verification mode, the person` identity is verified on the basis of known biometrics features. The system at IITK has been developed incorporating four traits like face, fingerprint, iris and signature. The system identifies the person by extracting features corresponding to four traits and the matching them against the stored database features, the result is computed on the basis of fusion of individual results to a final matching score. It is being tested on the IITK database.
In Identification mode, the system identifies the person on the basis of unknown input sample. This system is under the process of development incorporating four traits like face, fingerprint, iris and signature; however the efforts are made for efficient database indexing and searching thus reducing the search space. The expected outcome will be the top n% match.
Multimodal System being developed at IIT Kanpur
Prof. Phalguni Gupta
Department of Computer Science & Engineering
Indian Institute of Technology Kanpur