Skip to main content

View full-text article in PMC

. 2018 May 10;18(5):1501. doi: 10.3390/s18051501

Table 5.

Comparison of the proposed method with previous methods using MICHE-I dataset based on NICE-I evaluation protocol.

Method		$E_{a}$
Hu et al. [30]	Sub-dataset by iPhone5	0.0193
Hu et al. [30]	Sub-dataset by Galaxy S4	0.0192
Arsalan et al. [45]	Sub-dataset by iPhone5	0.00368
	Sub-dataset by Galaxy S4	0.00297
	Sub-dataset by Galaxy Tab2	0.00352
SegNet-Basic [57]	Sub-dataset by iPhone5	0.0025
	Sub-dataset by Galaxy S4	0.0027
	Sub-dataset by Galaxy Tab2	0.0029
Proposed IrisDenseNet	Sub-dataset by iPhone5	0.0020
	Sub-dataset by Galaxy S4	0.0022
	Sub-dataset by Galaxy Tab2	0.0021