A new accurate noise-removing approach for non-cooperative iris recognition
- 397 Downloads
The applications of biometric technology for automated personal identification become ubiquitous. Iris recognition is well known for high accuracy and reliability among the biometric traits. This paper presents an efficient noise-removing approach for non-cooperative iris recognition systems. Proposed method removed the noise factors including eyelids, eyelashes, reflections, out of framework, pupil and sclera. The novelty is to detect eyelashes and reflections through finding appropriate thresholds using a procedure called statistical decision making. The eyelids are detected using parabolic Hough transform in normalized iris image to increase computational speed. In addition, a coarse-to-fine strategy for accurate and fast iris localization is proposed. The Gabor-wavelet and a novel encoding strategy proposed in our previous work are also used here to generate the iris codes. We elaborate the principle of mask code generation to assign noisy bits in an iris code to exclude them in matching step. Experimental results on CASIA-IrisV3-Interval database show superiority of the proposed scheme among other state-of-the-art methods available in the literature.
KeywordsIris segmentation Noise-removing Statistical decision making Iris code Mask code Hough transform
We highly appreciate Iran Research Centre of Intelligent Signal Processing (RCISP) for its support to this research as a part of M.Sc. thesis.
- 5.Masek, L.: Recognition of human iris patterns for biometric identification. Bachelor thesis, University of Western Australia. http://www.csse.uwa.edu.au/~k/studentprojects/libor/ (2003)
- 6.Mahlouji, M., Noruzi, A.: Human iris segmentation for iris recognition in unconstrained environments. Int. J. Comput. Sci. Issues 9, 149–155 (2012)Google Scholar
- 9.Qichuan, T., Xirong, L., Ziliang, L., Linsheng, L.: Imperfect iris information for identity recognition. In: International Conference on Image and Signal Processing (2009)Google Scholar
- 10.Kong, W.K., Zhang, D.: Accurate iris segmentation based on novel reflection and eyelash detection model. In: International Symposium on Intelligent Multimedia, Video& Speech Processing (2001)Google Scholar
- 11.He, Z., Tan, T., Sun, Z., Qiu, X.: Robust eyelid, eyelash and shadow localization for iris recognition. In: Proceedings of the International Conference on Image Processing, pp. 265–268 (2008) Google Scholar
- 16.Aligholizadeh, M.J., Javadi, SH., Sabbaghi-Nadooshan, R., Kangarloo, K.: An effective method for eyelashes segmentation using wavelet transform. In: International Conference on Biometrics Kansei Engineering (2011)Google Scholar
- 18.Ghodrati, H., Dehghani, M.J., Danyali, H.: Iris feature extraction using optimized Gabor wavelet based on multi objective genetic algorithm. In: International Symposium on Innovations in Intelligent Systems and Applications (INISTA) (2011)Google Scholar
- 19.CASIA iris image database. http://www.cbsr.ia.ac.cn. Accessed 20 Oct 2008