Table 3 Performance of model fusion with different combinations in SkinFLNet.

	WP	WR	WF	WS	WSP
InceptionResNetV2 + InceptionV3	0.85	0.82	0.82	0.82	0.93
InceptionResNetV2 + ResNet50	0.85	0.82	0.82	0.82	0.93
InceptionResNetV2 + VGG16	0.82	0.80	0.80	0.80	0.91
InceptionResNetV2 + VGG19	0.84	0.79	0.79	0.79	0.93
InceptionResNetV2 + MobileNet	0.84	0.83	0.83	0.83	0.93
InceptionResNetV2 + MobileNetV2	0.84	0.80	0.80	0.80	0.93
InceptionV3 + Res50	0.85	0.82	0.82	0.82	0.93
InceptionV3 + VGG16	0.84	0.82	0.82	0.82	0.93
InceptionV3 + VGG19	0.85	0.79	0.79	0.79	0.94
InceptionV3 + MobileNet	0.84	0.81	0.81	0.81	0.93
InceptionV3 + MobileNetV2	0.85	0.80	0.80	0.80	0.93
ResNet50 + VGG16	0.83	0.81	0.81	0.81	0.93
ResNet50 + VGG19	0.86	0.81	0.81	0.81	0.94
ResNet50 + MobileNet	0.84	0.81	0.81	0.81	0.93
ResNet50 + MobileNetV2	0.86	0.82	0.82	0.82	0.93
VGG16 + VGG19	0.83	0.77	0.77	0.77	0.94
VGG16 + MobileNet	0.84	0.80	0.80	0.80	0.93
VGG16 + MobileNetV2	0.84	0.80	0.80	0.80	0.93
VGG19 + MobileNet	0.85	0.79	0.79	0.79	0.94
VGG19 + MobileNetV2	0.85	0.78	0.78	0.78	0.94
MobileNet + MobileNetV2	0.85	0.78	0.78	0.78	0.94

WP weight precision, WR weight recall, WF weight F-score, WS weight sensitivity, WSP weight specificity.
Significant values are in bold.

Quick links

Search