Acta Photonica Sinica, Volume. 54, Issue 3, 0330001(2025)
Super-resolution Solar Spectral Irradiance Reconstruction Method Based on Convolutional Neural Network
Figures & Tables(9)
Fig. 2. Comparison of magnitude，first derivative，and second derivative of the original spectrum h and two reconstructions spectra h1 and h2
Fig. 5. HSRS spectra and convolutional solar spectra and the response function at 443 nm of TSIS-1 SIM
Fig. 7. 0.1 nm resolution reconstruction of solar spectral irradiance
Table 1. The used solar spectra
View tableView in ArticleTable 1. The used solar spectra
Data product Spectral resolution Uncertainty TSIS-1 SIM 0.25~42 nm 0.24%~0.41% TSIS-1 HSRS 0.1 nm 0.3%~1.3%
Table 2. TSIS-1 SIM data product reconstruction quality
View tableView in ArticleTable 2. TSIS-1 SIM data product reconstruction quality
Method RMSE MAPE SAM PSNR SSIM Bandwidth correction 0.222 4 6.219 8 0.082 3 25.620 2 0.482 2 Janssen iteration 0.208 1 5.496 5 0.077 0 26.190 4 0.485 9 CNN 0.009 9 0.636 6 0.002 1 47.270 9 0.970 9
Tools
Get Citation
Copy Citation Text
Peng ZHANG, Jianwen WENG, Qing KANG, Jianjun LI. Super-resolution Solar Spectral Irradiance Reconstruction Method Based on Convolutional Neural Network[J]. Acta Photonica Sinica, 2025, 54(3): 0330001
Paper Information
Category:
Received: Sep. 3, 2024
Accepted: Nov. 27, 2024
Published Online: Apr. 22, 2025
The Author Email: Jianwen WENG (wengjw@aiofm.ac.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20