Study on the Relative Radiometric Calibration Method Based on HY-1E Satellite PMRIS

The Ocean Satellite HY-1E carries a programmable medium resolution imaging spectrometer (PMRIS), which consists of 5 camera sub-modules with a 14.4° field of view, and images by stitching the field of views of the 5 modules. Due to the response differences between the charge-coupled device (CCD) detectors, there are lots of vertical stripes which seriously affects subsequent quantitative applications. We propose an adaptive histogram matching method based on grayscale classification, which effectively eliminates the stripes in the PMRIS images. The comparison experiments are implemented using the laboratory relative radiometric calibration (RRC), life image statistics (LIS) RRC and moment matching (MM) method. After using the method we proposed, for land scenes, the maximum streaking coefficient of all sensors is smaller than 0.0067%, the average streaking coefficient is smaller than 0.0008%, and the median streaking coefficient is smaller than 0.0007%; For ocean scenes, the maximum streaking coefficient is smaller than 0.0032%, the average streaking coefficient is smaller than 0.0004%, and the median streaking coefficient is smaller than 0.0003%. The experiment shows that the proposed method can effectively improve the RRC quality of the HY-1E PMRIS images.