Different annulus regions of iris texture have various distribution characteristics. All the previous feature extraction methods are unable to make a difference between relevance of intra-annulus feature and difference of inter-annulus feature. With an analysis of relevance of intra-annulus, this paper proposes a kind of feature extraction method based on texture regions. The method firstly uses 2D-Gabor filter to independently extract and encode texture features from different regions respectively, and then the set of feature vectors are applied to classification and recognition by SVM classifier. The experimental results show that the proposed method has quite high recognition accuracy.
展开▼
