High Spectral Resolution Image Restoration Method Based on Filtered Phase Reconstruction
ZHENG Lian-hui1, XIE Yun2
1. College of Artificial Intelligence, Putian University, Putian 351100, China
2. Putian 95 Hospital, China Rongtong Healthcare Group Co., Ltd., Putian 351100, China
Abstract：A High spectral resolution spectrometer system has been widely applied in the fields of astronomical detection, remote sensing imaging, and target identification due to its high precision and resolution in spectral information acquisition. Through spectral image reconstruction, grating spectra can be decomposed into two-dimensional slit images and one-dimensional spectral data. However, in practical applications, factors such as wavefront aberrations and environmental noise significantly degrade imaging quality, limiting the system's performance in complex scenarios. In particular, during slit imaging, wavefront aberrations lead to a reduction in spatial resolution, which in turn affects the accuracy of subsequent spectral analysis and data interpretation. To address these issues, this paper proposes a high spectral image restoration method based on filtered phase reconstruction. The method first employs a Hartmann-Shack wavefront sensor to measure the wavefront aberration within the optical system, with a root mean square (RMS) measurement error of 0.002 μm when compared to interferometric results. Based on optical principles, the filtering effect of the slit on the wavefront aberration is analyzed under different slit widths, laying the foundation for subsequent phase reconstruction. Based on this analysis, a wavefront phase filtering model suitable for slit imaging conditions is established, enabling accurate estimation of the point spread function (PSF) and providing a physical and mathematical foundation for image restoration. In the image restoration stage, non-blind image restoration techniques are applied to reconstruct high-quality slit images using the estimated PSF. Both numerical simulations and experimental results demonstrate that the proposed method effectively restores image details even under challenging imaging conditions such as large wavefront aberrations and low signal-to-noise ratios, significantly improving the spatial resolution and spectral stability of the imaging system. Compared to conventional methods that rely on adaptive optics (AO) systems for real-time correction, the proposed approach does not require additional hardware, offering greater flexibility and improved engineering feasibility. Furthermore, the experimental results also demonstrate that the method performs robustly across a wide spectral range and even under white light illumination, thereby further expanding the applicability of high spectral imaging technology in fields such as astronomical observation and ground-based remote sensing. In conclusion, the proposed image restoration method offers an effective technical solution for enhancing the performance of high-spectral-resolution spectral imaging systems under complex optical conditions, demonstrating strong practical value and broad application prospects.
[1] Chelpanov A A, Kobanov N I, Kolobov D Y. Solar Physics, 2016, 291(11): 3329.
[2] Korsos M B, Georgoulis M K, Gyenge N, et al. The Astrophysical Journal, 2020, 896(2): 119.
[3] Fischer C E, Vigeesh G, Lindner P, et al. The Astrophysical Journal Letters, 2020, 903(1): L10.
[4] Obridko V N, Fainshtein V G, Zagainova Y S, et al. Solar Physics, 2020, 295: 149.
[5] Zwaan C. Annual Review of Astronomy & Astrophysics, 1987, 25(1): 83.
[6] Chae J, Park H M, Ahn K, et al. Solar Physics, 2013, 288(1): 1.
[7] Hölken J, Doerr H P, Feller A, et al. Astronomy & Astrophysics, 2024, 687: A22.
[8] Oba T, Shimizu T, Katsukawa Y, et.al. Solar Physics, 2022, 297(9): 114.
[9] Luo J, Nie Y, Ren W, et al. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024, 46(8): 5541.
[10] Hampson K M, Turcotte R, Miller D T, et al. Nature Reviews Methods Primers, 2021, 1: 68.
[11] Bai H C, Yang P, Zhao L M, et al. The Astrophysical Journal, 2023, 956(1): 62.
[12] Rao C H, Gu N T, Rao X J, et al. Science China Physics, Mechanics &Astronomy, 2020, 63: 109631.
[13] ZHAO Si-min, GU Nai-ting, HUANG Lin-hai, et al(赵思旻, 顾乃庭, 黄林海, 等). Infrared and Laser Engineering(红外与激光工程), 2023, 52(7): 20220887.
[14] Zheng L H, Rao C H, Gu N T, et al. Optics Express, 2016, 24(1): 153.
[15] ZHENG Lian-hui, RAO Chang-hui, GU Nai-ting, et al(郑联慧, 饶长辉, 顾乃庭, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2016, 36(12): 4088.
[16] Zheng L H, Xie Y. Journal of Optical Technology, 2019, 86(10): 634.
[17] ZHENG Lian-hui, GU Nai-ting, RAO Chang-hui, et al(郑联慧, 顾乃庭, 饶长辉，等). Acta Optica Sinica(光学学报), 2015, 35(8): 0812002.
[18] Zhong L B, Tian Y, Rao C H. Optics Express, 2014, 22(23): 29249.
[19] Jia P, Lü J M, Ning R Y, et al. Monthly Notices of the Royal Astronomical Society, 2024, 527(3): 6581.
[20] Hua X, Pan C, Shi Y, et al. IEEE Transactions on Geoscience and Remote Sensing, 2022, 60: 1.
[21] Jayavel A, Gopinath S, Augamuthu P P, et al. Photonics, 2023, 10(4): 396.
[22] Gu S, Zhang L, Zuo W, et al. Proceedings of the 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014, 2862.
[23] Danielyan A, Katkovnik V, Egiazarian K. IEEE Transactions on Image Processing, 2011, 21(4): 1715.
[24] Zha Z, Yuan X, Zhou J, et al. IEEE Transactions on Image Processing, 2020, 29: 8561.
[25] Ma T H, Huang T Z, Zhao X L, et al. Information Sciences, 2017, 408: 213.
[26] Ren D, Zuo W, Zhang D, et al. IEEE Transactions on Image Processing, 2018, 27(1): 511.
[27] Zeng H, Cao J, Zhang K, et al. 2024 IEEE Conference on Computer Vision and Pattern Pecognition, 2024, 27820.
[28] Perrone D, Favaro P. 2014 IEEE Conference on Computer Vision and Pattern Recognition, 2014, 2909.
[29] Dong J X, Roth S, Schiele B. Transactions on Pattern Analysis and Machine Intelligence, 2022, 44(12): 9960.
[30] Praveen P A, Arockiaraj F G, Gopinath S, et al. Photonics, 2022, 9(9): 625.
[31] Anand V, Han M, Maksimovic J, et al. Opto-Electronis Science, 2022, 1(3), 210006.
[32] Gopinath S, Angamuthu P P, Kahro T, et al. Photonics, 2023, 10(1): 3.
