Cervical Lesion Classification Method Based on Cross-Validation Decision Fusion Method of Vision Transformer and DenseNet

Table 4

Comparison fusion results of different models.

Model	ω1	ω2	ACC
MobileNetV3 + ShuffleNetV2	0.65	0.35	0.61
Vision transformer + ShuffleNetV2	0.9	0.1	0.61
Vision transformer + MobileNetV3	0.3	0.7	0.61
ShuffleNetV2 + EfficientNetV2	0.1	0.9	0.65
MobileNetV3 + DenseNet161	0.3	0.7	0.66
Vision transformer + EfficientNetV2	0.05	0.95	0.66
MobileNetV3 + EfficientNetV2	0.35	0.65	0.67
ShuffleNetV2 + DenseNet161	0.1	0.9	0.67
EfficientNetV2 + DenseNet161	0.4	0.6	0.68
Vision Transformer + DenseNet161	0.05	0.95	0.68