Research On Iris Recognition Algorithm

As an emerging biometric for identification, iris recognition has received increasing attention in recent years. One of the most promising recognition techniques is the one based on the human iris due to its high reliability for personal identification. It has desirable properties such as uniqueness, stability, noninvasiveness and living-object character over the traditional method. To overcome the deficiencies of the existing recognition algorithms, this thesis has studied the iris image preprocessing and the feature matching.his image preprocessing plays a key role in iris recognition. It will greatly affect the feature extraction and matching in the following process. To overcome such disadvantages as larger memory space and time cost of the traditional preprocessing methods, a preprocessing method combining intensity curve and Hough transform has been proposed in this thesis. It is simple and efficient.In iris feature extraction, Fourier-wavelet transform in two dimension spaces is used in the thesis. The method keeps translation, rotation and scale-invariant, but dimension of extracting feature is very high. Therefore, a resulting high-dimensional feature vector is mapped into a low-dimensional feature subspace using improving Fisher linear discriminant. The results show that the method reduces dimension of features and gets effective classification. And it can improve accurate rate of recognition.Iris recognition algorithm has been implemented by using Matlab language and simulated in the CASIA iris image database. The algorithm can achieve a good result.