An Efficient and Robust Iris Segmentation Algorithm Using Deep Learning

<div>Manually labeled region information of the iris. (a) The manually labeled iris region. (b) The bounding box computed from the circle parameters of the limbus boundary is the region label for training Faster R-CNN. (c) The information of the pupillary region used for training GMM.</div>

Mobile Information Systems

fig9

Figure 9

Figure 9: An Efficient and Robust Iris Segmentation Algorithm Using Deep Learning 