Table 5.

Comparison of MCCBAM with other attention mechanisms.

Attention mechanism	Grading	Precision	Recall	F1-score	Accuracy
SAM	I	0.769	0.714	0.741	0.754
	II	0.698	0.714	0.706
	III	0.795	0.833	0.814
SENet	I	0.696	0.929	0.796	0.762
	II	0.806	0.595	0.685
	III	0.821	0.762	0.790
SKNet	I	0.878	0.857	0.867	0.833
	II	0.761	0.833	0.795
	III	0.872	0.810	0.840
Non_Local	I	0.857	0.857	0.857	0.841
	II	0.846	0.786	0.815
	III	0,0.822	0.881	0.851
CBAM	I	0.850	0.810	0.829	0.817
	II	0.786	0.786	0.786
	III	0.818	0.857	0.837
BAM	I	0.923	0.857	0.889	0.833
	II	0.767	0.786	0.776
	III	0.818	0.857	0.837
MCCBAM	I	0.902	0.881	0.892	0.873
	II	0.850	0.810	0.829
	III	0.867	0.929	0.897