The Potential Application of Innovative Methods in Neural Networks for Surface Crack Recognition of Unshelled Hazelnut

Table 4

The effect of the number of filters at CNN layers on the accuracy of DCNN models.

Model	Number of filters at CNN layers	Dropout	Accuracy (%)
Proposed model	16-32	—	94.08
	16-32-64	—	97.76
	16-32-64	0.5	97.89
	16-32-64-128	—	97.32
	16-32-64-128-256	—	97.40