Table 4.

Comparisons of the proposed method with previous methods using NICE-II dataset.

Method	${\mathit{E}}_{\mathit{a}}$
Luengo-Oroz et al. [69]	0.0305
Labati et al. [70]	0.0301
Chen et al. [71]	0.029
Jeong et al. [26]	0.028
Li et al. [72]	0.022
Tan et al. [73]	0.019
Proença et al. [74]	0.0187
de Almeida [75]	0.0180
Tan et al. [76]	0.0172
Sankowski et al. [77]	0.016
Tan et al. [35]	0.0131
Haindl et al. [78]	0.0124
Zhao et al. [79]	0.0121
Arsalan et al. [45]	0.0082
SegNet-Basic [57]	0.00784
Proposed IrisDenseNet	0.00695