The application of Convolutional Neural Networks to Detect Slow, Sustained Deformation in InSAR Timeseries
This work addresses the problem of automated global monitoring for volcanic and urban environments, representing an incremental improvement by applying CNNs to InSAR data with specific optimizations.
The study tackled detecting slow, sustained deformation in InSAR timeseries using Convolutional Neural Networks, achieving detection thresholds as low as 1.8cm for deformation signals and improving classification performance by up to 15% with over-wrapping in real-world tests.
Automated systems for detecting deformation in satellite InSAR imagery could be used to develop a global monitoring system for volcanic and urban environments. Here we explore the limits of a CNN for detecting slow, sustained deformations in wrapped interferograms. Using synthetic data, we estimate a detection threshold of 3.9cm for deformation signals alone, and 6.3cm when atmospheric artefacts are considered. Over-wrapping reduces this to 1.8cm and 5.0cm respectively as more fringes are generated without altering SNR. We test the approach on timeseries of cumulative deformation from Campi Flegrei and Dallol, where over-wrapping improves classication performance by up to 15%. We propose a mean-filtering method for combining results of different wrap parameters to flag deformation. At Campi Flegrei, deformation of 8.5cm/yr was detected after 60days and at Dallol, deformation of 3.5cm/yr was detected after 310 days. This corresponds to cumulative displacements of 3 cm and 4 cm consistent with estimates based on synthetic data.