Boosting Quality of Pansharpened Images Using Deep Residual Denoising Network

Bin Yang; Xiang Wang

doi:10.3788/LOP56.161009

Laser & Optoelectronics Progress, Volume. 56, Issue 16, 161009(2019)

Boosting Quality of Pansharpened Images Using Deep Residual Denoising Network

Bin Yang^1,2、* and Xiang Wang^1,2

Author Affiliations

¹ School of Electrical Engineering, University of South China, Hengyang, Hunan 421001, China

² Hunan Provincial Key Laboratory for Ultra-Fast Micro/Nano Technology and Advanced Laser Manufacture, University of South China, Hengyang, Hunan 421001, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(12)

Fig. 1. Framework of proposed method

Download full size

Fig. 2. Structure of boosting network

Download full size

Fig. 3. Experimental images. (a) MS; (b) PAN

Download full size

Fig. 4. Results of different fusion methods. (a) ATWT; (b) BT; (c) GIHS; (d) GS; (e) SVT; (f) SWT

Download full size

Fig. 5. Boosted results of different fusion methods. (a) ATWT; (b) BT; (c) GIHS; (d) GS; (e) SVT; (f) SWT

Download full size

Fig. 6. Influences of different network parameters on experimental results. (a) ERGAS; (b) SAM; (c) Q4; (d) CC

Download full size

Fig. 7. Reference image and results of compared methods and proposed method. (a) Reference image; (b) CS; (c) DRPNN; (d) PNN; (e) proposed method

Download full size

Fig. 8. Reference image and results of compared methods and proposed method. (a) Reference image; (b) CS; (c) DRPNN; (d) PNN; (e) proposed method

Download full size

Fig. 9. Different reference images and results of proposed method. (a) Reference images; (b) results of proposed method

Download full size

Table 1. Running time of different methods and corresponding boosting stages
View table
Table 1. Running time of different methods and corresponding boosting stages
Method ATWT BT GIHS GS SWT SVT
Fusion 0.0327 0.0542 0.0323 0.1653 1.1115 0.1728
Boosting 1.0147 0.9499 0.9509 0.9479 0.9540 0.9546
Total 1.0474 1.0042 0.9832 1.1132 2.0655 1.1274

Table 2. Evaluation of fusion results obtained by different methods and corresponding boosted results

View table

Table 2. Evaluation of fusion results obtained by different methods and corresponding boosted results

Index	ATWT		BT		GIHS
Index	Fused result	Boosted result	Fused result	Boosted result	Fused result	Boosted result
ERGAS	5.5413	3.6464	7.7943	3.5788	5.5766	3.8262
SAM	8.5281	5.7892	6.0326	5.4821	6.5539	5.7464
Q4	0.7085	0.8444	0.7600	0.8616	0.7479	0.8590
CC	0.8487	0.9358	0.7822	0.9375	0.8481	0.9297
Index	GS		SVT		SWT
Index	Fused result	Boosted result	Fused result	Boosted result	Fused result	Boosted result
ERGAS	4.7565	3.2772	4.0625	2.8333	4.5673	3.2977
SAM	5.8587	5.2991	6.1018	4.5629	6.9649	5.2726
Q4	0.7900	0.8875	0.8035	0.9044	0.7837	0.8727
CC	0.9117	0.9490	0.9223	0.9614	0.8988	0.9478

Table 3. Evaluation of compared methods and proposed method

View table

Table 3. Evaluation of compared methods and proposed method

Figure	Index	CS	DRPNN	PNN	Proposed method	Ideal
Fig. 7	ERGAS	3.1382	3.2030	2.4409	2.3934	0
	SAM	4.5610	3.7884	3.7599	3.6627	0
	Q4	0.7631	0.7447	0.8032	0.8005	1
	CC	0.9510	0.9532	0.9705	0.9698	1
Fig. 8	ERGAS	2.5521	2.7164	2.2530	2.2412	0
	SAM	3.4794	3.1021	3.3527	3.3413	0
	Q4	0.6290	0.6229	0.6463	0.6338	1
	CC	0.9612	0.9587	0.9725	0.9676	1
Fig. 9(The 1st column)	ERGAS	3.7382	3.6131	3.0033	2.5599	0
	SAM	5.8313	5.0033	5.0254	4.2870	0
	Q4	0.8693	0.8565	0.9095	0.9169	1
	CC	0.9353	0.9371	0.9568	0.9684	1
Fig. 9(The 2nd column)	ERGAS	3.9656	4.2478	3.1395	3.0194	0
	SAM	6.1953	5.2592	5.2617	4.9211	0
	Q4	0.8917	0.8562	0.9333	0.9303	1
	CC	0.9285	0.9149	0.9538	0.9572	1
Fig. 9(The 3rd column)	ERGAS	3.7939	3.6033	2.8936	2.1136	0
	SAM	5.8606	4.6764	4.6806	3.6695	0
	Q4	0.8656	0.8472	0.9142	0.9377	1
	CC	0.9329	0.9350	0.9585	0.9777	1

Tools

Get Citation

Copy Citation Text

Bin Yang, Xiang Wang. Boosting Quality of Pansharpened Images Using Deep Residual Denoising Network[J]. Laser & Optoelectronics Progress, 2019, 56(16): 161009

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites