A Markov Random Field and Adaptive Regularization Embedded Level Set Segmentation Model Solving by Graph Cuts

Table 3

Region overlap metrics of different algorithms corresponding to Figure 6 in the same order.

Image number	Algorithms	JS	DSC	FPR	FNR
Image #1	GAC	0.0033	0.0067	0.1095	0.9967
	LSR	0.0026	0.0053	0.3212	0.9974
	CV	0.0034	0.0069	0.1387	0.9966
	FTC	0.0031	0.0061	0.1266	0.9845
	LBF	0.0030	0.0059	0.1971	0.9970
	LAC	0.0031	0.0062	0.1533	0.9969
	LPFE	0.0023	0.0045	0.3869	0.9977
	NFT	0.0012	0.0024	0.7226	0.9988
	Our method	0.8919	0.9429	0.0365	0.0769
Image #2	GAC	0.0109	0.0215	0.0310	0.9891
	LSR	0.0085	0.0169	0.0303	0.9915
	CV	0.0090	0.0178	0.0207	0.9910
	FTC	0.0119	0.0236	0.0205	0.9881
	LBF	0.0099	0.0196	0.0455	0.9901
	LAC	0.0117	0.0232	0.0038	0.9883
	LPFE	0.0123	0.0242	0.0076	0.9877
	NFT	0.0076	0.0151	0.0265	0.9924
	Our method	0.8853	0.9392	0.0034	0.0573
Image #3	GAC	0.0973	0.1773	0.0807	0.9027
	LSR	0.1055	0.1908	0.0088	0.8944
	CV	0.1120	0.2014	0.3003	0.8880
	FTC	0.0948	0.1732	0.4070	0.9052
	LBF	0.1001	0.1820	0.0066	0.8998
	LAC	0.0959	0.1750	0.0019	0.9041
	LPFE	0.0914	0.1676	0.0016	0.9085
	NFT	0.1124	0.2021	0.0082	0.8875
	Our method	0.9580	0.9786	0.0010	0.0209