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

Table 2

The fusion performance of the seven methods and the proposed method from Figures 9–11.

Method	Flower				Cup				Calendar
	SD	Q^AB/F	En	MI	SD	Q^AB/F	En	MI	SD	Q^AB/F	En	MI
PCNN	40.61	0.41	3.90	4.72	38.22	0.43	4.01	4.21	37.42	0.40	4.88	4.31
Contourlet	40.62	0.40	4.02	4.73	38.39	0.42	4.03	4.38	37.65	0.52	5.35	4.47
Shearlet	40.81	0.42	4.06	4.89	38.44	0.45	4.08	4.41	38.15	0.56	5.55	4.55
GAN	40.81	0.57	4.80	4.99	39.10	0.51	4.10	4.63	38.59	0.60	5.76	4.91
CSR	40.89	0.59	4.65	4.77	38.55	0.54	4.10	4.60	39.03	0.66	5.90	4.89
SR-SML	41.19	0.57	4.66	5.12	38.71	0.55	4.14	4.72	39.11	0.67	5.93	4.91
DCNN	40.16	0.60	4.81	5.16	40.01	0.58	4.09	4.81	39.98	0.65	5.98	5.24
Proposed	41.26	0.68	4.96	5.27	40.21	0.63	4.18	4.93	39.91	0.67	6.00	5.29