Spatial Resolution Enhancement Technology of FY-3D MWRI Based on Convolutional Auto-Encoder

During data collection, the Microwave Radiation Imager (MWRI) on the FY3D satellite is affected by various degradation factors, resulting in varying degrees of reduced spatial resolution of multi-channel data. Irreversible high-frequency information degradation occurs during the degradation process. However, the existing spatial resolution enhancement methods cannot perfectly restore the high-frequency information of the original measurement. In this study, an algorithm based Convolutional Auto-Encoder(CAE) is proposed to improve the image resolution of MWRI and restore high-frequency information. Experimental results on simulated and actual MWRI images have demonstrated the validity and effectiveness of the method.