Iris recognition based on Empirical Mode Decomposition

Empirical mode decomposition (EMD), a multi-resolution decomposition technique, is adaptive and suitable for nonlinear, non-stationary data analysis. We adopt the EMD approach to decompose the iris images into many ingredients with different frequency range, and exploit the mutual information criterion to extract proper parts of them as features for recognition. The proposed method can solve the problem of illumination variation and eyelid-eyelash occlusion. Experimental results show that the performance of the proposed method is encouraging and comparable to the state-of-the-art iris recognition algorithm.