The Fusion of Multi-Focus Images Based on the Complex Shearlet Features-Motivated Generative Adversarial Network

Table 1

The fusion performance of the seven methods and the proposed method from Figures 6–8.

Method	Pepsi-Cola				Clock				Plane
	SD	Q^AB/F	En	MI	SD	Q^AB/F	En	MI	SD	Q^AB/F	En	MI
PCNN	43.71	0.36	6.58	4.46	39.27	0.50	6.98	5.01	44.97	0.40	3.91	3.39
Contourlet	44.11	0.59	7.00	5.39	39.41	0.57	7.00	5.19	46.24	0.48	3.98	3.35
Shearlet	44.07	0.61	7.10	5.34	39.44	0.52	7.00	5.21	46.70	0.49	4.06	3.41
GAN	44.25	0.70	7.16	5.38	39.95	0.62	7.06	5.20	45.842	0.69	4.04	3.72
CSR	45.23	0.76	7.10	5.50	40.50	0.68	7.03	5.42	48.10	0.73	4.08	3.60
SR-SML	45.40	0.78	7.11	5.55	40.88	0.69	7.05	5.44	48.90	0.76	4.19	3.62
DCNN	44.80	0.74	7.06	5.41	39.66	0.65	7.00	5.32	46.85	0.68	4.03	3.58
Proposed	45.25	0.78	7.20	5.60	40.88	0.69	7.10	5.53	50.15	0.76	4.28	3.72