Iris Recognition Using Semantic Indexing

Abstract

The iris of human eye is one of the most useful traits for biometric recognition. This paper presents an iris recognition system based on semantic indexing. The proposed system uses the concepts of latent semantic indexing (LSI) for iris recognition. One technique of LSI is the singular value decomposition (SVD). The SVD is an information retrieval uses numerical decomposition methods to compute one characteristic value (i.e. SVD) for each iris image to be used as a recognition feature. The proposed system consists of two phases: the training and recognition. The training phase is responsible on storing the iris models in the database, while the task of recognition phase is to compute the similarity measure between the SVD of the query iris image and SVDs of the iris images found in the database. The recognition decision is made according to the normalized similarities and appeared as a text message tells what the identity it is. The successful recognition rate was about 96%, which ensure the successful of the employed method and correct path of computations.