Remote Sensing Images Mosaicking Method Based on Spatiotemporal Fusion

Chaoqi He; Qize Li; Hualin Liu; Jingbo Wei

doi:10.3788/LOP202158.1415002

Journals >Laser & Optoelectronics Progress >Volume 58 >Issue 14 >Page 1415002 > Article

Laser & Optoelectronics Progress
Vol. 58, Issue 14, 1415002 (2021)

Remote Sensing Images Mosaicking Method Based on Spatiotemporal Fusion

Chaoqi He, Qize Li, Hualin Liu, and Jingbo Wei^*

Author Affiliations

School of Information Engineering, Nanchang University, Nanchang, Jiangxi 330031, China

show less

DOI: 10.3788/LOP202158.1415002 Cite this Article Set citation alerts

Chaoqi He, Qize Li, Hualin Liu, Jingbo Wei. Remote Sensing Images Mosaicking Method Based on Spatiotemporal Fusion[J]. Laser & Optoelectronics Progress, 2021, 58(14): 1415002 Copy Citation Text

show less

Fig. 1. Process of spatiotemporal fusion method

Download full size

Fig. 2. Cascaded convolutional neural network for spatiotemporal fusion

Download full size

Fig. 3. Integral mosaic results of different algorithms. (a) ENVI algorithm; (b) ENBLEND algorithm; (c) ENFUSE algorithm; (d) QP algorithm; (e) MOST method

Download full size

Fig. 4. Local area of Fig. 3. (a) ENVI algorithm; (b) ENBLEND algorithm; (c) ENFUSE algorithm; (d) QP algorithm; (e) MOST method; (f) LandSat

Download full size

Fig. 5. Mosaic results at different reference moments. (a) LandSat (December); (b) MOST (October); (c) LandSat (October); (d) MOST (November); (e) LandSat (November)

Download full size

Fig. 6. Mosaic results of different algorithms in transition region. (a) ENVI algorithm; (b) ENBLEND algorithm; (c) ENFUSE algorithm; (d) QP algorithm; (e) MOST method

Download full size

Block	Algorithm	RMSE	CC	SSIM	MI	SAM	ERGAS	RASE	Q4
Block 1	ENVI	68	0.986	0.973	4.622	0.141	0.158	0.150	0.983
	ENBLEND	3	1.000	1.000	4.847	0.003	0.007	0.007	1.000
	ENFUSE	113	0.891	0.704	1.099	0.064	0.244	0.238	0.868
	QP	--	0.987	0.896	4.883	0.050	0.311	0.310	0.893
	MOST	2	1.000	1.000	5.217	0.001	0.004	0.003	1.000
Block 2	ENVI	219	0.673	0.861	1.501	0.073	0.516	0.490	0.601
	ENBLEND	210	0.694	0.875	1.646	0.063	0.497	0.473	0.626
	ENFUSE	126	0.817	0.701	0.830	0.062	0.300	0.290	0.781
	QP	--	0.903	0.873	1.698	0.076	0.339	0.334	0.868
	MOST	166	0.778	0.843	1.301	0.059	0.391	0.380	0.738
Block 3	ENVI	132	0.902	0.879	1.858	0.076	0.307	0.294	0.865
	ENBLEND	83	0.939	0.943	2.093	0.047	0.199	0.194	0.930
	ENFUSE	103	0.891	0.869	1.081	0.058	0.256	0.247	0.858
	QP	--	0.934	0.782	2.055	0.140	0.487	0.481	0.775
	MOST	91	0.923	0.909	1.496	0.054	0.229	0.227	0.913
Block	Algorithm	RMSE	CC	SSIM	MI	SAM	ERGAS	RASE	Q4
Total	ENVI	153	0.854	0.900	1.857	0.096	0.357	0.340	0.818
	ENBLEND	130	0.868	0.941	2.266	0.038	0.302	0.289	0.845
	ENFUSE	114	0.870	0.788	1.000	0.061	0.267	0.259	0.842
	QP	--	0.936	0.852	2.242	0.090	0.391	0.388	0.838
	MOST	109	0.898	0.920	1.955	0.039	0.257	0.250	0.885

Table 1. Digital evaluation of mosaicking results

Chaoqi He, Qize Li, Hualin Liu, Jingbo Wei. Remote Sensing Images Mosaicking Method Based on Spatiotemporal Fusion[J]. Laser & Optoelectronics Progress, 2021, 58(14): 1415002

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information