A Complementary Augmentation Algorithm for Remotely Sensed Data Based on Radiometer Distortion Phenomenon

For the problem of distortion and loss of radiometer observation time series data, LSTM-based Seq2Seq network is used to augment and supplement the contours, and CNN network and RNN network are used as comparison items. Firstly, the basic principles network is introduced respectively; secondly, the microwave remote sensing data in Harbin are used to conduct data augmentation experiments, and finally the relevant conclusions are analyzed.