Fruit Classification Model Based on Residual Filtering Network for Smart Community Robot

Table 4

Comparison with different features using RF.

	Accuracy (Avg)	Recall (Avg)	F1 score (Avg)	Precision (Avg)
RFN-RF	98.898%	98.899%	98.852%	98.848%
RGB-RF	90.015%	89.902%	89.744%	89.893%
CNN-RF	97.948%	97.914%	97.865%	97.945%