Multi-unit iris biometric fusion using gray level co-occurrence matrix features

Iris offers an excellent recognition performance when used as a biometric. This is because no two irises are alike, not between identical twins, or even between the left and right eye of the same individual. Irises are also stable; unlike other identifying characteristics that can change with age, the pattern and textural details of a individual's iris is fully formed by ten months of age and remains the same for the duration of his lifetime. This paper proposes multi-unit biometric fusion recognition system. In this paper we have fused matching scores from left and right iris of a person using the gray level co-occurrence matrix (GLCM) for textural feature extraction. From the proposed fusion framework there has been significant improvement in the performance compared to Unimodal iris recognition system. The proposed fusion method has been tested using CASIA-iris-V4 thousand database.