InSAR-Simulator

InSAR Data Simulator for generating synthetic InSAR noisy images with ground truth clean phase and cohrerence estimation.

This Simulator is used for paper "DeepInSAR: A Deep Learning Framework for SAR Interferometric Phase Restoration and Coherence Estimation". Please check out the branch 'DeepInSAR-v1' to get the same version which is used in paper's experiments.

Generate a set of 2D signals

python gen_2d_signals.py \
   seed=1234 \
   samples=500 \
   workers=5 \
   height=512 \
   width=512 \
   processing_dir=/mnt/hdd1/3vG_data/sim/sim2d/2021-03-21-500_500/