Applied Informatics

Table 1 Qualitative comparison of the results of our method based on MSWS and MCS on the ten images presented in Fig. 10

From: Object segmentation by saliency-seeded and spatial-weighted region merging

No.	PRI		GCE		VoI
No.	MCS	MSWS	MCS	MSWS	MCS	MSWS
38,092	0.6229	0.6979	0.0810	0.0604	2.4606	2.4236
41,033	0.7001	0.7010	0.0293	0.0272	1.7014	1.6997
62,096	0.5739	0.5962	0.2916	0.0138	1.9264	1.227
101,087	0.3914	0.4014	0.0326	0.0277	2.9608	2.9295
108,082	0.7099	0.7292	0.0701	0.0568	1.2445	1.1788
123,074	0.3353	0.3389	0.0337	0.0236	2.2989	2.2529
160,068	0.6084	0.6487	0.1227	0.0584	2.0237	1.9315
175,043	0.8036	0.8116	0.0277	0.0107	1.111	1.1087
296,059	0.6776	0.6784	0.0293	0.0256	2.0147	1.9971
376,043	0.6378	0.6475	0.0271	0.0162	1.704	1.6399

The average values over 300 images of the BSDS300 dataset are also included
Italic indicates best performance

Back to article page