Iris Feature Extraction using Wavelet Transform

S. Minhas, M.Y. Javed, and A. Basit(Pakistan)


Feature extraction, Hamming distance, Iris recognition, Wavelet transforms.


Using biometric technology, a number of ways are provided to fulfill the today’s enormous security needs. Iris recognition is one of the biometric technologies and proves to be the most dependable and accurate method for person identification. Iris feature extraction is the critical stage of the recognition process and is the main focus of this paper. In this paper, the performance of various feature extraction techniques is evaluated. These include wavelet transforms using several variations of daubechies, biorthogonal, symlets and coiflets wavelets. Different experiments are performed using CASIA version 1 iris database. All the methods are applied using different types of feature vectors to extract the iris information. By comparing the feature vectors using hamming distance, it is found that the best method provides an accuracy of 99.83%.

Important Links:

Go Back