The Triplet Network Enhanced Spectral Diversity (T-NESD) Method for the Correction of TOPS Data Co-registration Errors for Non-Stationary Scenes

In this work, a novel approach for the correction of misregistration errors in sequences of Terrain Observation with Progressive Scan (TOPS) Sentinel-1 SAR data is presented. The method represents a further evolution of the Enhanced Spectral Diversity (ESD) approaches. Remarkably, the developed algo...

Full description

Saved in:
Bibliographic Details
Main Authors: Mastro, Pietro, Pepe, Antonio
Format: Conference Proceeding
Language:English
Subjects:
Azimuth
Co-registration errors
Diversity methods
Earthquakes
Electrostatic discharges
Geoscience and remote sensing
large ground deformations
SAR
Strain
Synthetic aperture radar
TOPSAR
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this work, a novel approach for the correction of misregistration errors in sequences of Terrain Observation with Progressive Scan (TOPS) Sentinel-1 SAR data is presented. The method represents a further evolution of the Enhanced Spectral Diversity (ESD) approaches. Remarkably, the developed algorithm is almost insensitive to the presence of large azimuth ground displacements due, for instance, to massive earthquakes, volcanic eruptions or glacier movements. Indeed, in such non-stationary contexts, the conventional ESD and network ESD approaches for the SAR TOPS data co-registration reveals problematic being co-registration errors and azimuth ground deformation components mixed out. Preliminary experiments conducted on a set of TOP SAR data related to the area hit by the Ridgecrest earthquake MW 7.1, California, on July 04 2019 confirm the validity of the theoretical framework.
ISSN:2153-7003
DOI:10.1109/IGARSS47720.2021.9554439