A Novel Deep Convolutional Neural Network Based on ResNet-18 and Transfer Learning for Detection of Wood Knot Defects

Table 8

Prediction results of ReSENet-18 model and other classical CNN models.

Different models	Number	Accuracy	Parameters (million)
LeNet-5	495	81.01%	0.06
AlexNet	586	95.91%	60
VGGNet-16	576	94.27%	138
GoogLeNet	557	91.16%	6.8
MobileNet V2	580	94.93%	3.5
ReSENet-18 (our method)	605	99.02%	33