Towards more accurate iris recognition system by using hybrid approach for feature extraction along with classifier

Arif Ullah, Abdu Salam, Hanane El Raoui, Dorsaf Sebai, Mahnaz Rafie

Abstract


Iris recognition become one of the most accurate and reliable steadfast human biometric recognition system of the decad. This paper presents an accurate framework for iris recognition system using hybrid algorithm in preprocess and feature extraction section. The proposed model for iris recognition with significant feature extraction was divided into three main levels. First level is having pre-processing steps which are necessary for the desired tasks. Our model deploys on three types of datasets such as UBIRIS, CASIA, and MMU and gets optimal results for performing activity. At last, perform matching process with decision based classifier for iris recognition with acceptance or rejection rates. Experimental based results provide for analysis according to the false receipt rate and false refusal amount. In the third level, the error rate will be checked along with some statistical measures for final optimal results. Constructed on the outcome the planned method provided the most efficient effect as compared to the rest of the approach.

Keywords


Classifier; Extraction; Feature; Hybrid; Iris

Full Text:

PDF


DOI: http://doi.org/10.11591/ijres.v11.i1.pp59-70

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

View IJRES Stats