Intelligent Processing and Applications of Optical Remote Sensing Data from Fengyun Satellites (Invited)

Chuyao Luo; Xu Huang; Jiazheng Li; Xutao Li; Yunming Ye

doi:10.3788/AOS241175

Acta Optica Sinica, Volume. 44, Issue 18, 1800006(2024)

Intelligent Processing and Applications of Optical Remote Sensing Data from Fengyun Satellites (Invited)

Chuyao Luo¹, Xu Huang², Jiazheng Li², Xutao Li², and Yunming Ye^2、*

Author Affiliations

¹School of Electronics and Information Engineering, Harbin Institute of Technology, Shenzhen 518055, Guangdong , China

²School of Computer Science and Technology, Harbin Institute of Technology, Shenzhen 518055, Guangdong , China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(14)

Fig. 1. Schematic diagram of IR-MAD^[6]

Download full size

Fig. 2. Structure diagram of CNN calibration model^[7]

Download full size

Fig. 3. Calibration results of different methods. (a) Original image; (b) calibration results with MST; (c) calibration results with IR-MAD; (d) calibrate results with CNN deep learning method

Download full size

Fig. 4. Overview of cloud detection methods for remote sensing satellites

Download full size

Fig. 5. Framework of STCD-UNet^[29]

Download full size

Fig. 6. Multi-scale time conditional generative adversarial network^[60]

Download full size

Fig. 7. Overview of precipitation retrieval algorithms based on satellite observations

Download full size

Fig. 8. Vision Transformer-based precipitation retrieval^[83]

Download full size

Fig. 9. Sea ice image segmentation network (MSDA-Net)^[100]

Download full size

Fig. 10. Model performance in different representative scenarios

Download full size

Table 1. Results of intelligent radiation self-calibration experiment

View table

Table 1. Results of intelligent radiation self-calibration experiment

Sensor	Method	Degradation (absolute difference to MST) /%
Sensor	Method	Band 1	Band 2	Band 6	Band 7	Band 8	Band 9	RMSE
FY-3A VIRR	MST	16.96 (-)	12.74 (-)	7.18 (-)	33.77 (-)	26.98 (-)	22.61 (-)	-
	IR-MAD^[6]	15.73 (1.24)	10.94 (1.80)	5.18 (2.00)	33.22 (0.55)	28.37 (1.40)	23.41 (0.80)	1.39
	CNN^[7]	16.57 (0.40)	12.33 (0.41)	5.73 (1.45)	32.06 (1.71)	27.30 (0.32)	22.31 (0.30)	0.96
FY-3B VIRR	MST	23.71 (-)	16.53 (-)	4.52 (-)	35.48 (-)	27.54 (-)	27.62 (-)	-
	IR-MAD^[6]	21.66 (2.05)	14.51 (2.02)	4.38 (0.14)	37.54 (2.05)	29.87 (2.33)	27.65 (0.03)	0.73
	CNN^[7]	22.69 (1.02)	16.96 (0.43)	5.51 (0.99)	35.37 (0.12)	28.00 (0.47)	26.70 (0.92)	0.74
FY-3C VIRR	MST	16.69 (-)	14.89 (-)	7.24 (-)	19.47 (-)	12.10 (-)	14.73 (-)	-
	IR-MAD^[6]	11.83 (4.86)	10.26 (4.63)	5.53 (1.71)	18.33 (1.14)	11.84 (0.26)	12.18 (2.55)	3.05
	CNN^[7]	12.48 (4.23)	12.00 (2.89)	6.16 (1.08)	18.83 (0.64)	11.79 (0.31)	11.65 (3.08)	2.50

Table 2. Comparison of experimental results on MERSI dataset

View table

Table 2. Comparison of experimental results on MERSI dataset

Method	Jaccard	Precision	Recall	F1-score	MIoU	Acc
FCN^[30]	87.1	94.3	91.9	93.1	88.2	93.8
U-Net^[31]	88.8	95.5	92.8	94.1	89.8	94.7
DeepLabv3+^[32]	87.3	94.2	92.3	93.2	88.4	93.9
SegFormer^[33]	84.9	92.8	90.8	91.8	86.1	92.6
Swin-Unet^[34]	91.0	95.1	95.4	95.2	91.7	95.8
ConvUNeXt^[35]	90.6	95.6	95.2	95.4	91.4	95.6
GCDB-UNet^[36]	91.1	97.3	93.5	95.4	91.9	95.8
STCD-UNet (ours)	92.7	96.2	96.2	96.2	93.3	96.6

Table 3. Results of satellite cloud image extrapolation methods

View table

Table 3. Results of satellite cloud image extrapolation methods

Model	256×256				512×512				Average
Model	$↑$ PSNR	$↓$ MSE	$↓$ MAE	$↓$ GDL	PSNR	MSE	MAE	GDL	PSNR	MSE	MAE	GDL
ConvLSTM^[46]	25.00	324.39	12.64	4.78	24.28	388.63	13.98	4.64	24.64	356.51	13.31	4.71
TrajGRU^[47]	25.01	341.37	12.93	4.72	23.26	604.79	16.10	5.73	24.14	473.08	14.52	5.23
PredRNN^[48]	25.06	337.12	12.84	4.86	24.12	410.62	14.67	4.64	24.59	373.87	13.75	4.75
PredRNN++^[49]	24.75	337.38	13.33	4.84	24.30	369.76	14.09	4.69	24.53	353.57	13.71	4.77
MIM^[50]	24.67	329.35	13.02	4.84	24.10	375.16	14.18	4.70	24.39	352.26	13.60	4.77
PhyDNet^[53]	24.21	327.41	13.08	4.86	23.26	388.52	13.88	4.93	23.74	357.97	13.48	4.90
SA-ConvLSTM^[52]	24.08	366.74	13.99	13.63	23.31	449.29	15.72	13.74	23.70	408.02	14.86	13.69
MSTCGAN^[60]	25.41	280.91	11.06	4.68	24.33	326.25	12.96	4.53	24.87	303.58	12.01	4.61

Table 4. Comparison of indicators for different semantic segmentation models

View table

Table 4. Comparison of indicators for different semantic segmentation models

Model	MIoU	FWloU	Acc	Jaccard	Precision	Recall	Fl-score
U-Net^[31]	80.1	81.9	88.7	85.4	89.0	95.4	92.2
ResUNet^[101]	81.6	82.4	89.3	85.6	90.1	94.5	92.3
DenseUNet^[102]	82.8	83.6	89.8	86.7	90.1	95.9	93.0
DeepLabv3^[94]	89.9	90.4	94.3	92.2	94.4	97.5	96.0
DeepLabv3+^[32]	85.3	86.0	91.6	88.5	92.1	95.8	94.0
TransUNet^[103]	85.8	86.5	91.6	89.2	91.2	97.6	94.3
Swin-Unet^[34]	86.2	86.8	91.9	89.3	91.9	97.0	94.5
DAU-Net^[96]	89.2	89.7	94.1	91.4	94.8	96.2	95.5
SegFormer^[33]	89.6	90.1	94.0	91.9	94.1	97.5	95.8
ConvUNeXt^[35]	88.1	88.6	93.6	90.5	93.9	95.1	95.0
VAN^[104]	89.7	90.2	94.3	91.9	94.7	96.9	95.8
MSDA-Net (ours)	93.0	93.4	96.1	94.5	96.3	98.1	97.2

Tools

Get Citation

Copy Citation Text

Chuyao Luo, Xu Huang, Jiazheng Li, Xutao Li, Yunming Ye. Intelligent Processing and Applications of Optical Remote Sensing Data from Fengyun Satellites (Invited)[J]. Acta Optica Sinica, 2024, 44(18): 1800006

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: Reviews

Received: Jun. 17, 2024

Accepted: Aug. 21, 2024

Published Online: Sep. 11, 2024

The Author Email: Ye Yunming (yeyunming@hit.edu.cn)

DOI:10.3788/AOS241175

CSTR:32393.14.AOS241175

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology