Performance comparison of full 2-D DCT, 2-D Walsh and 1-D transform over row mean and column mean for iris recognition

In this paper we propose a novel iris recognition method which reduces the computational complexity and increases the accuracy. Iris recognition enjoys universality, high degree of uniqueness and moderate user co-operation. This makes Iris recognition systems unavoidable in emerging security & authentication mechanisms. Iris recognition is one of the important techniques and is rotation invariant. In this paper we have tested full 2-dimensional Discrete Cosine Transform (DCT), full 2-dimensional Walsh Transform (WHT), and the proposed method DCT/WHT row mean and column mean. Row mean DCT/WHT gives the best performance with the accuracy of 75.78% outperforming full 2-dimensional DCT/WHT with low accuracy around 66.10% further proposed Walsh Row/Column mean requires 99.96% less computations as that of full 2-D DCT. Thus our proposed method not only gives better accuracy but also reduces computational time considerably.