A Review on Biometric Authentication Using Adaptive Iris Features

There are several biometric identities associated with the human which prominently includes the fingerprint, palm-print, palm-vein, finger-vein, retina and Iris features. The human beings are identified by their biological identities for the attendance systems, authorization systems or other similar applications. The biometric systems have found their way into almost all of the organizations with the medium to large employee base. Many of the small organizations with the adequately higher number of employees are also incorporating the biometric systems. The biometric systems based upon the Iris features are being popular as the standalone or hybrid biometric or with other authentication entities. The Iris recognition requires the accurate localization of the Iris features from the image of eye collected for training or testing purposes. The Iris extraction requires two demarcation circles, where first circle demarcates the outer boundary and second circle demarcates the inner boundary by detecting the outer boundary of the pupil. Also, the angular shift mechanism can be incorporated to study the movement of the Iris in the given image for accurate localization of the region of interest containing the Iris feature. The proposed solution will utilize the probabilistic classification based upon the multi-class SVM to detect the Iris features with or without contact lenses. The proposed solution aims at improving the existing model for the robust performance.