Research on Spectral Interferogram Bidirectional Correction Method for Spatial Heterodyne Remote Sensing Data
Wang Xin-Qiang,Wang Zhen,Liang Qiu-Yu,Xiong Wei,Li Zhi-Wei,Ye Song,Gan Yong-Ying,Wang Fang-Yuan
DOI: https://doi.org/10.3964/j.issn.1000-0593(2024)04-0963-06
2024-01-01
Spectroscopy and spectral analysis
Abstract:Spatial Heterodyne Spectroscopy(SHS),a new hyperspectral analysis technology,has been widely used in atmospheric detection,satellite remote sensing and other fields.However,because the fabrication of spatial heterodyne spectroscopy is not ideal or the change of working environment will change the instrument parameters and introduce errors,the interferogram data is not accurate,so error correction is needed.Due to the huge difference between the satellite platform and the ground environment,the correction parameters measured on the ground are no longer applicable to the spatial heterodyne interferogram data,especially the change of modulation errors(phase errors and non-uniform errors),which greatly affects the accuracy of the spectrum.Based on spatial heterodyne modulation of the remote sensing data error,from two aspects of spectrum and the interference figure,separation and analysis of the causes of error,think mainly comes from the CCD size scale spatial heterodyne remote sensing data error caused by spectral frequency change and the CCD response to a change in intensity of interference pattern change,spectral interferogram bidirectional correction method is proposed.Twelve O2 absorption spectra measured by the Greenhouse Gases Monitoring Instrument(GMI)on GF-5 were selected for calibration.Onewas taken as the calibration spectrum,and the error-free spectrum simulated by SCIATRAN was compared with the calibration spectrum.The frequency deviation of the two spectra in the spectral dimension was analyzed,and the frequency of the characteristic peak determined the frequency transformation relationship between the two.Then,the simulated spectrum is stretched in frequency according to the transformation relationship so that the simulated spectrum after stretching coincides with the measured spectrum peak.The interferogram of the stretching simulation spectrum and the measured spectrum is calculated,and the changing relationship of the interferogram intensity is obtained by comparing the interferogram of the two.Finally,the intensity variation relation of the interferogram is used to correct the other 11 spectra,and the corrected spectra are obtained.To measure the correction effect,The standard deviation(STD),mean square error(MSE)and signal-to-noise ratio(SNR)of the corrected spectra were calculated;results show that both STD and MSE were significantly lower,with a significant increase in SNR,the basic and STD are below 0.07,the SNR can reach more than 20.The STD of the spectrum with the best correction effect decreased by 0.376 7,SNR increased by 25.101 6,and MSE decreased by 0.158 7.The STD of the spectrum with poor correction effect decreased by 0.229 6,SNR increased by 9.632 8,and MSE decreased by 0.104 9.To sum up,it shows that the spectral interferogram bidirectional correction method proposed in this paper has a good effect on error correction of spatial heterodyne remote sensing data,and the processing process is simple,providing a new direction for similar data processing.