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

Table 3

Experimental environment.

Hardware environment		Software environment
Memory	16.00GB	System	Windows 10
CPU	Intel Core i5-4210H 2.90GHz (2 core)	Environment configuration	Pytorch-gpu 1.0.0 + Python 3.7.3 + cuda 8.0 + cudnn7.1.3
Graphics card	NVIDIA GeForce GTX 960 M(2G)