Table 6.

Comparison with other fine-tuned models based on pre-trained deep learning models using average classification accuracy (%) and training parameters (in millions) on three datasets (D1, D2, and D3). Bold emphasis indicate the best results

Method	D1(%)	D2(%)	D3(%)	Params
Incep.-V3, 2016 [46]	65.55	83.44	80.95	26
ResNet50, 2016 [35]	62.24	74.15	67.58	36.6
DenseNet121, 2017 [47]	64.61	80.40	75.86	11
Incep.-ResnetV2, 2017 [48]	68.10	83.93	84.35	57
MobileNet, 2017 [49]	67.33	82.35	84.16	7
EfficientNetB0, 2019 [39]	56.03	81.82	72.87	12
Ours (VGG-16)	79.58	85.43	87.49	18
Ours (VGG-19)	74.84	82.83	85.00	21.2