Detection and Localization of Iris Features for Person Identification using Digital Image Processing Techniques and Pattern Recognition Approach
Ashwini Chate, Pramod Kumar, Sushilkumar N. Holambe · International Journal of Computer Applications · 2021
Biometrics Systems are computerized methods for verification or recognition to the identity of a living person on the basis of some physiological characteristics.Human eyes are completely unique and can be used for biometric purposes.In the proposed research work, extraction of iris is done by using digital image processing techniques.For experimentation and implementation, a dataset of IIT Delhi, UBIRIS and UPOL iris image database has been used and the evaluation results clearly demonstrate that the proposed iris segmentation technique provides better accuracy in iris recognition rather than existing techniques.Classification is done by using supervised and unsupervised pattern classification techniques.Support vector machine and k-means clustering is used for classification iris features.Overall got 97.25% accuracy based on confusion matrix.The results clearly prove that the proposed technique provides better, rather than existing techniques.