Iris texture recognition based on multilevel 2-D Haar wavelet decomposition and hamming distance approach

Security has played an essential role in human life and it has been motivated the governments of all regions in the world to make security measures tight by a higher level of safety. Although major progresses are carried out in the iris recognition domain, it will have continuous to be an open issue in the future and it will deserve additional investigation. So, as to refer these difficulties, in this paper a novel iris recognition method for feature extraction is proposed based on 2-D Haar wavelet decomposition approach. It is chosen because it causes to reduce the level of noises of the iris texture effectively. Furthermore, it accelerates the extraction process of the iris pattern and it simplifies the computing process. In order to implement this method, multilevel 2-D wavelet decomposition approach is applied on the iris images which are normalized in the pre-processing step and then, it extracts the features of these images to create a distinctive code. For finding similarity among the two iris images and carrying out the matching process, we use Hamming distance measure in order to obtain high accuracy rate in our proposed iris recognition system. Our experimental results on CASIA-Iris V3 database show the effectiveness of the proposed method in iris recognition system.