A Novel and Efficient Method for Iris Automatic Location

Evaluating iris segmentation for scenario optimisation

“…Many new segmentation methods [5][6][7][8][9][10][11][12][13] have been published in recent years to address these factors and to improve the performance of iris segmentation for more robust iris recognition systems. In this paper, however, in order to be consistent across the reported experiments and comparisons, we investigate only segmentation algorithms for which performance has been reported based on the same sample database, in this case the CASIA V1.0 database [14].…”

“…Xu et al [10] form a block in the binarised eye image which includes the pupil, and then the determined block parameters are transformed to form a pupil boundary. Subsequently, the iris boundary is detected by applying a proposed edge detector to rectangular areas which are transformed from two parts of the iris.…”

“…In order to localise the iris boundary, a linear filter is proposed and applied to rectangular areas which were transformed from two parts of the iris as in [10].…”

“…For our comparative study, results reported in the more recently proposed methods investigated [5][6][7][8][9][10][11][12]15] are utilized.…”

“…In this paper, a particular goal includes conducting a detailed comparative study of the two earliest and most influential segmentation methods [2] and [3] with several more recently proposed methods [5][6][7][8][9][10][11][12]. We also include as an important component in our study a comparison with our own recently proposed segmentation method [4].…”