Green Fluorescent Protein and Phase-Contrast Image Fusion via Generative Adversarial Networks

Table 1

Objective evaluation of different fusion methods.

Metric	Statistics	DTCWT	CVT	NSCT	SR	CNN	SFL-CT	CST	Proposed
QMI	MV	0.2916	0.2725	0.3070	0.3877	0.4233	0.5142	0.5396	0.6362
	SD	0.0555	0.0493	0.0578	0.0461	0.1647	2.5361	25.4793	0.0438
	WT	0	0	0	0	16	9	12	111
QP	MV	0.4374	0.4076	0.5011	0.6056	0.5218	0.5303	0.5597	0.7098
	SD	0.1645	0.1550	0.1616	0.1382	0.1881	0.1799	0.1750	0.0697
	WT	0	0	0	3	18	1	11	115
QY	MV	0.7185	0.7066	0.7372	0.7087	0.8570	0.8520	0.8938	0.8925
	SD	0.0743	0.0739	0.0701	0.0746	0.1298	0.1001	0.1060	0.0669
	WT	0	0	2	0	16	0	75	57
QCB	MV	0.4540	0.4541	0.4634	0.4491	0.4646	0.4885	0.4991	0.4908
	SD	0.1045	0.1030	0.1081	0.1095	0.1350	0.1363	0.1377	0.1275
	WT	0	8	11	0	8	5	69	47
VIF	MV	0.6697	0.6623	0.6968	0.6926	0.7368	0.6785	0.7705	0.8682
	SD	0.1362	0.1343	0.1323	0.0869	0.1678	0.1147	0.1290	0.1236
	WT	0	0	0	0	29	0	29	90

The abbreviations MV, SD, and WT stand for mean value, standard deviation, and winning times, respectively.