SVD and DWT Based Iris Recognition Using Beagleboard-xM

Iris recognition is one of the most accurate biometric methods in use today. However, the iris recognition algorithms are currently implemented on general purpose sequential processing systems, such as generic central processing units (CPUs). One of the remaining problems in iris recognition is the implementation of its efficient hardware systems. This proposed work uses a hardware Beagleboard-xM, which is designed specifically to address the Open Source Community. Unlike regular iris recognition systems, the proposed method also involves pre-processing, normalization, feature extraction and matching. Here the iris normalization is performed using a new method called Circular Masking technique. And for feature extraction, Biorthogonal Wavelet Transform and Singular Value Decomposition (SVD) algorithms are used. And Euclidean distance measure is used in order to find similarity between the iris images. And the results demonstrate significant improvements in iris recognition accuracy. And all these operations are implemented on Beagleboard-xM using OpenCV library functions. Keywords— Circular Masking, Biorthogonal Wavelet, SVD, Euclidean Distance, Beagleboard-xM, OpenCV