O. Nomir and E. Radwan (Egypt)
Biometrics, human identification, iris, segmentation, Haar transform, hamming distance.
Iris has always played a very important role in human identification. The biomedical literature suggests that irises are as distinct as fingerprints, and DNA. In this research work, we develop a human identification system based on iris features. In this system, the identification is carried out by analyzing and comparing a given human iris image against a database of iris records to find the best matches. We archive the irises records in a database. The system automatically manipulates the archived irises images, extracts and archives representative feature vectors for each iris image. The features are extracted using Haar transform. These extracted features are later used for retrieval. Given an individual’s iris image, the proposed system extracts its representative features and retrieves the best matches from the database based on the minimum Hamming distance. The experimental results show that system is robust and using iris feature in the human identification is an effective solution for biometric identification.
Important Links:
Go Back
