Sclera Vessel Analysis for Biometric System on Real Time Security Application

Biometrics has an unavoidable space in the era of security in terms of identifying the Individuals. There are many biometrics which are providing more prominent result, but all these biometrics needs the user cooperation in acquisition. The sclera, white portion of the eye, vessel patterns are holding the uniqueness to identify the individuals. This becomes the root of many researchers to recognize the individual with non-co-operative user images. Sclera recognitions are beginning of the new era which is to leads the recognition on the surveillance camera. Since the Image quality is less in the surveillance camera images and the segmentation of sclera is more challenging which has more impact on the accuracy of the recognition. To improve the accuracy of the segmentation Convolutional Neural Network has been used which has increase the time complexity of the system. Since the accuracy of the segmentation has increased and to optimize the time integro Differential Operator is utilized which reduces the searching area of the iris so the time taken will be reduced. The combination of CNN and DIO will increase the accuracy and complexity of the Segmentation Process. The experiments are carried out on the UBIRIS Dataset V2.