An OCaNet Model Based on Octave Convolution and Attention Mechanism for Iris Recognition

Table 4

Comparison of performance between the single modal network and multimodal network.

Model	Accuracy (%)	Number of misjudgments
ResNet50	84.34	515
OCa–ResNet50	85.35	482
MobileNet	87.87	399
OCa–MobileNet	89.85	334
ResNet18	86.07	458
OCaNet (ours)	91.52	279