INTERNATIONAL JOURNA L OF ENGINEERING SCI ENCES & RESEARCH TECHNOLOGY Efficient iris Recognition System to I mprove Performance M easure for live videos
D. Suganya · 2013
Iris recognition is regarded as the most reliable a nd accurate biometric identification system availab le. Iris ation is one of the most successful applications in video analysis and image processing. In this proje ct, several proficient approaches are proposed to incre asing the performance measures of Iris Recognition system for irst and foremost step, efficient face detection al gorithm such as Viola algorithm is used for this purpose. Using feature e xtraction process left and right eye are isolated a nd stored in the database. Next is the Iris Segmentation process. Ir is segm entation is almost the most challenging part in iri s recognition. The main purposes are to improve accur acy and to reduce computational time of iris locali zation. Briefly, this approach tries to explore regions of interests (ROI) among image regions and to localize iris from one or more remaining regions. A new eyelash detection algorithm based on direction filter to achieve a lo w rate of eyelash misclassification is proposed. Final step i s the Iris Recognition, in order to speed up the re sponse time and to enhance the code searching speed Euclidean Distance Classifier is used to efficiently recognize the ir is images from