An iris segmentation using harmony search algorithm and fast circle fitting with blob detection

Pupil and iris segmentation based on ellipsis or circle recognition are sensitive to light reflections reflected images. The method presented here is independent of size shape at the same time insensitive mirror pupil detected using algorithm can be a reference point further sclera eye as well iris. also effective when not positioned perpendicularly camera eye. algorithm’s average accuracy for all tested databases was 96% considering only noisy distorted images whilst result 100% achieved with unblurred clear proposed quickly simply reproduced combination known image processing methods. developed detecting eyelid boundaries poor quality due use edge approximation Harmony Search Algorithm. An optimized detection used detect its edges. A variation eliminate shadows eyelashes. scheme UBIRIS.v1 database, MMU.v1 database MILES databases, providing high results short time. Segmentation 98.14%, – 90% 99.8%.