An iris recognition method based on zigzag collarette area and asymmetrical support vector machines

We propose an improved iris recognition method for person identification using an iris segmentation approach based on chain code and zigzag collarette area with support vector machine (SVM). The zigzag collarette area is selected as a personal identification pattern which captures only the most important areas of iris complex pattern and better recognition accuracy is achieved. The idea to use the zigzag collarette area is that it is insensitive to the pupil dilation and usually not affected by eyelids or eyelashes. The deterministic feature sequence is extracted from iris images using Gabor wavelet technique and used to train SVM as iris classifiers. The traditional SVM is modified as asymmetrical SVM to treat False Accept and False Reject differently to satisfy several security requirements. The parameters of SVM are tuned to improve overall system performance. Our experimental results also indicate that the performance of SVM as a classifier is far better than the performance of backpropagation neural network (BPNN), K-nearest neighbor (KNN), Hamming and Mahalanobis distance. The proposed innovative technique is computationally effective as well as reliable in term of recognition rate of 99.56%.