English

Generative Adversarial Networks for Synthesizing InSAR Patches

Signal Processing 2020-08-05 v1 Machine Learning Image and Video Processing

Abstract

Generative Adversarial Networks (GANs) have been employed with certain success for image translation tasks between optical and real-valued SAR intensity imagery. Applications include aiding interpretability of SAR scenes with their optical counterparts by artificial patch generation and automatic SAR-optical scene matching. The synthesis of artificial complex-valued InSAR image stacks asks for, besides good perceptual quality, more stringent quality metrics like phase noise and phase coherence. This paper provides a signal processing model of generative CNN structures, describes effects influencing those quality metrics and presents a mapping scheme of complex-valued data to given CNN structures based on popular Deep Learning frameworks.

Keywords

Cite

@article{arxiv.2008.01184,
  title  = {Generative Adversarial Networks for Synthesizing InSAR Patches},
  author = {Philipp Sibler and Yuanyuan Wang and Stefan Auer and Mohsin Ali and Xiao Xiang Zhu},
  journal= {arXiv preprint arXiv:2008.01184},
  year   = {2020}
}

Comments

accepted in preliminary version for EUSAR2020 conference

R2 v1 2026-06-23T17:36:58.605Z