Table 1.

Simulation testing: comparing the ARI of DCBFC with the other clustering methods using different picture size.

Image size	DCBFC	K-means	Hierarchical	Spectral	PCA-k-means	PCA-hierarchical	Spectral-threshold
$64\times 64$	$0.94\pm 0.08$	$0.76\pm 0.10$**	$0.95\pm 0.07(p=0.989)$	$0.81\pm 0.03$**	$0.74\pm 0.08$**	$0.90\pm 0.10(p=0.050)$	$0.89\pm 0.10$**
$128\times 128$	$0.99\pm 0.02$	$0.80\pm 0.09$**	$0.99\pm 0.01(p=1.000)$	$0.80\pm 0.07$**	$0.75\pm 0.08$**	$0.99\pm 0.01(p=1.000)$	$0.92\pm 0.09$*
$256\times 256$	$0.99\pm 0.01$	$0.83\pm 0.07$**	$0.99\pm 0.00(p=1.000)$	$0.84\pm 0.12$**	$0.77\pm 0.08$**	$0.99\pm 0.00(p=1.000)$	$0.94\pm 0.07$*

Note: Two-way ANOVA tests with FDR correction were used to compare the significant difference between ARI of DCBFC and the other methods.

^*$p<0.01$.

^**$p<0.001$.