Reconstruction of Missing Data in Time-Series of Optical Satellite Images Using Self-Organizing Kohonen Maps

摘要

The paper offers the reconstruction method of data in time-series of optical satellite images of high and medium spatial resolution. The method is based on self-organizing Kohonen maps (SOMs). The reconstruction accuracy was analyzed on data of satellites Landsat-8 (30 m) and Sich-2 (8 m). The average error of reconstruction was 11 % for Landsat-8 and 4 % for Sich-2.