Explainable Artificial Intelligence-Based IoT Device Malware Detection Mechanism Using Image Visualization and Fine-Tuned CNN-Based Transfer Learning Model

Table 1

Comparison of detection accuracy between four different CNN pretrained models.

Pretrained model	Total parameters	Total depth	Accuracy (dataset 1)	Accuracy (dataset 2)
DenseNET201	18,520,103	707	0.947	0.757
MobileNET	3,337,383	87	0.942	0.764
ResNET50	23,798,631	175	0.868	0.387
Inception-v3	22,012,734	311	0.969	0.786