Multi-Modal Data Fusion for Land-Subsidence Image Improvement in PSInSAR Analysis

There are three popular methods to understand the land subsidence: leveling, Global Navigation Satellite System, and Interferometric Synthetic Aperture Radar (InSAR) analysis using SAR images. While both leveling System can measure amount of subsidence only at specific points, InSAR observe a wide area in short time intervals. In terms accuracy, however, is inferior leveling; centimeter/millimeter order (InSAR/PSInSAR analysis) vs. millimeter (leveling). Among all observation errors analysis, tropospheric delay error has large adverse effect on measurement. It difficult suppress this by conventional because they try remove each pixel independently an image. However, geometrically-neighboring regions/pixels should be naturally correlated. Our proposed method employs such neighboring relationship convolutional neural network (CNN). CNN designed improve mutually incorporating image error, which estimated any methods. Experimental results demonstrate that our reduce mean compared with method: from 10.3mm 6.80mm.