Anis Farihan Mat Raffei, Rohayanti Hassan, Shahreen Kasim, Hishamudin Asmuni, Asraful Syifaa' Ahmad, Rahmat Hidayat, Ansari Saleh Ahmar
The quality of eye image data become degraded particularly when the image is taken in the non-cooperative acquisition environment such as under visible wavelength illumination. Consequently, this environmental condition may lead to noisy eye images, incorrect localization of limbic and pupillary boundaries and eventually degrade the performance of iris recognition system. Hence, this study has compared several segmentation methods to address the abovementioned issues. The results show that Circular Hough transform method is the best segmentation method with the best overall accuracy, error rate and decidability index that more tolerant to 'noise' such as reflection. © 2018 Authors.
Faculty of Computing, Universiti Teknologi Malaysia, Skudai, Johor, Malaysia; Faculty of Computer Science and Information Technology, Universiti Tun Hussein Onn Malaysia, Johor, Malaysia; Department of Information Technology, Politeknik Negeri Padang, Indonesia; Department of Statistics, Universitas Negeri Makassar, Makassar, Indonesia; AHMAR Institute, Makassar, Indonesia
