Table 1.

Performance of different methods on the GBM dataset.

	Method	DictionarySize=256	DictionarySize=512	DictionarySize=1024
160 training	MCSCSPM-HE	92.71 ± 0.91	93.01 ± 1.10	92.65 ± 0.75
	MCSCSPM-RGB	92.58 ± 0.94	92.50 ± 0.86	92.47 ± 0.73
	MCSCSPM-Gray	86.33 ± 1.12	86.74 ± 0.91	86.69 ± 0.81
	PSDSPM [5]	91.02 ± 1.89	91.41 ± 0.95	91.20 ± 1.29
	SMLSPM [4]	92.35 ± 0.83	92.57 ± 0.91	92.91 ± 0.84
	ScSPM [23]	79.58 ± 0.61	81.29 ± 0.86	82.36 ± 1.10
	KSPM [17]	85.00 ± 0.79	86.47 ± 0.55	86.81 ± 0.45
80 training	MCSCSPM-HE	91.41 ± 1.07	91.19 ± 0.91	91.13 ± 0.93
	MCSCSPM-RGB	90.88 ± 1.06	91.28 ± 0.82	90.85 ± 0.67
	MCSCSPM-Gray	84.67 ± 1.63	84.53 ± 1.58	84.56 ± 1.62
	PSDSPM [5]	88.63 ± 0.91	88.91 ± 1.18	88.64 ± 1.08
	SMLSPM [4]	90.82 ± 1.28	90.29 ± 0.68	91.08 ± 0.69
	ScSPM [23]	77.65 ± 1.43	78.31 ± 1.13	81.00 ± 0.98
	KSPM [17]	83.81 ± 1.22	84.32 ± 0.67	84.49 ± 0.34
40 training	MCSCSPM-HE	89.16 ± 1.04	89.21 ± 0.75	88.84 ± 0.83
	MCSCSPM-RGB	89.24 ± 1.03	89.46 ± 1.14	89.53 ± 1.20
	MCSCSPM-Gray	81.37 ± 1.55	81.31 ± 1.19	80.80 ± 1.71
	PSDSPM [5]	84.06 ± 1.16	83.72 ± 1.46	83.40 ± 1.14
	SMLSPM [4]	88.05 ± 1.38	87.88 ± 1.04	88.54 ± 1.42
	ScSPM [23]	73.60 ± 1.68	75.58 ± 1.29	76.24 ± 3.05
	KSPM [17]	80.54 ± 1.21	80.56 ± 1.24	80.46 ± 0.56