A fuzzy approach to filtering interferometric SAR data

Abstract

A new method is proposed for interferometric synthetic aperture radar (SAR) data filtering. Considering the overestimation of the correlation map and the reliability of the phases of the master and slave images after registration, the authors propose a new fuzzy filter, based on the weighted least square method and the improved median filter. This method can be employed to filter noise and to strongly reduce the number of residues. Using Pi‐SAR X‐band data from a volcano in Kaimon, Japan, the advantages of the new approach are demonstrated.