Iris Liveness Detection

In recent years, iris recognition is becoming a very active topic in both research and practical applications. However, fake iris is a potential threat; there are potential threats for iris based systems. Iris Liveness Detection presents a classification of genuine and fake iris images based on dense Scale-Invariant Feature Transform (SIFT) and Linear Discriminant Analysis (LDA). Firstly, segmentation of the valid iris texture regions from the original iris images and normalization of the ring-shape iris regions into a unified coordinate system is done. Secondly, dense SIFT descriptors are extracted as the low level features for iris image classification to obtain the common components of texture primitives across different iris images. Then cluster the low level features using k-means and apply LDA for dimensionality reduction which takes advantages of unique dimensionality reduction in features compared to Hierarchical Visual Codebook (HVC). Lastly, Linear Support Vector Machine (SVM) classifier is employed to classify the genuine and fake iris images. Extensive experiments are conducted on a database containing fake and genuine iris images captured by iris devices. Experimental results show that the iris liveness detection can detect fake iris and genuine iris effectively.