Multi-Focus Image Fusion Method Based on Double-Scale Decomposition and Random Walk

Xiaomiao Li; Yanchun Yang; Jianwu Dang; Yangping Wang

doi:10.3788/LOP202259.2210011

Journals >Laser & Optoelectronics Progress >Volume 59 >Issue 22 >Page 2210011 > Article

Laser & Optoelectronics Progress
Vol. 59, Issue 22, 2210011 (2022)

Multi-Focus Image Fusion Method Based on Double-Scale Decomposition and Random Walk

Xiaomiao Li, Yanchun Yang^*, Jianwu Dang, and Yangping Wang

Author Affiliations

School of Electronic and Information Engineering, Lanzhou Jiaotong University, Lanzhou 730070, Gansu , China

show less

DOI: 10.3788/LOP202259.2210011 Cite this Article Set citation alerts

Xiaomiao Li, Yanchun Yang, Jianwu Dang, Yangping Wang. Multi-Focus Image Fusion Method Based on Double-Scale Decomposition and Random Walk[J]. Laser & Optoelectronics Progress, 2022, 59(22): 2210011 Copy Citation Text

show less

Fig. 1. Random walk model

Download full size

Fig. 2. Block diagram of proposed method

Download full size

Fig. 3. Decision diagram. (a) No small area removal; (b) small area removal; (c) guided filter decision diagram; (d) no guided filter fusion diagram; (e) guided filter fusion diagram

Download full size

Fig. 4. Experimental results. (a) Source 1; (b) source 2; (c) GFF; (d) SML; (e) GFDF; (f) NSST; (g) proposed method

Download full size

Fig. 5. Comparison of difference diagrams. (a) Source; (b) GFF; (c) SML; (d) GFDF; (e) NSST; (f) proposed method

Download full size

Image No.	Method	Q_IE	Q_AG	Q_G	Q_P	Q_Y	Q_FMI
Img1	GFF	7.0903	12.9282	0.7122	0.9100	0.9845	0.6093
	SML	7.0990	13.5688	0.7171	0.9132	0.9885	0.6168
	GFDF	7.0991	13.6048	0.7172	0.9141	0.9886	0.6172
	NSST	7.1216	13.2823	0.6286	0.8786	0.9308	0.4918
	Proposed method	7.0993	13.6062	0.7174	0.9139	0.9888	0.6174
Img2	GFF	7.1854	10.2186	0.6770	0.7841	0.9844	0.6031
	SML	7.1885	10.3361	0.6804	0.7840	0.9877	0.6140
	GFDF	7.1887	10.3420	0.6800	0.7840	0.9877	0.6144
	NSST	7.1369	8.6028	0.4923	0.6282	0.8670	0.3921
	Proposed method	7.1889	10.3431	0.6806	0.7844	0.9879	0.6147
Img3	GFF	7.0882	7.3046	0.7700	0.8503	0.9518	0.5911
	SML	7.0998	7.5253	0.7665	0.8632	0.9631	0.6241
	GFDF	7.1197	7.5816	0.7654	0.8709	0.9652	0.6417
	NSST	7.1098	7.1705	0.6968	0.6727	0.9139	0.5353
	Proposed method	7.1230	7.5964	0.7631	0.8733	0.9647	0.6472
Img4	GFF	7.7537	8.6300	0.7054	0.8619	0.9796	0.5647
	SML	7.7486	8.7299	0.7122	0.8633	0.9883	0.5952
	GFDF	7.7485	8.7288	0.7118	0.8636	0.9885	0.5963
	NSST	7.7722	7.1838	0.5345	0.6211	0.8719	0.3872
	Proposed method	7.7485	8.7287	0.7125	0.8642	0.9886	0.5964
Img5	GFF	7.8224	9.8855	0.6679	0.8913	0.9665	0.5732
	SML	7.8239	10.0865	0.6840	0.8923	0.9847	0.6038
	GFDF	7.8237	10.0993	0.6853	0.8927	0.9870	0.6091
	NSST	7.8137	8.7748	0.5166	0.6666	0.8701	0.4426
	Proposed method	7.8239	10.1002	0.6863	0.8930	0.9882	0.6102
Img6	GFF	7.8926	39.3042	0.6665	0.7099	0.9646	0.5469
	SML	7.8890	41.8240	0.6863	0.7227	0.9831	0.5650
	GFDF	7.8858	41.9490	0.6866	0.7248	0.9835	0.5701
	NSST	7.7600	39.2220	0.5422	0.4572	0.8715	0.4469
	Proposed method	7.8843	42.0410	0.6887	0.7280	0.9875	0.5737

Table 1. Comparison of indicators

Xiaomiao Li, Yanchun Yang, Jianwu Dang, Yangping Wang. Multi-Focus Image Fusion Method Based on Double-Scale Decomposition and Random Walk[J]. Laser & Optoelectronics Progress, 2022, 59(22): 2210011

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information