Complex-valued Wasserstein GAN for SAR Image Generation

Victor Dhédin (1) , Jérémie Levi (1) , Jérémy Fix (2, 3, 1) , Chengfang Ren (4) , I. Hinostroza (4)

1 CentraleSupélec
2 LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
3 BISCUIT - Bio-Inspired, Situated and Cellular Unconventional Information Technologies
4 SONDRA - Sondra, CentraleSupélec, Université Paris-Saclay

Complex-Valued (CV) Synthetic Aperture Radar (SAR) image generation and augmentation is an important pillar to enhance deep learning performance for SAR applications such as detection, classification, segmentation, super-resolution etc. Usual transformations (flip, rotation, translation, scaling, etc.) are mostly inapplicable to SAR images due to the radar characteristics and the processing pipeline. In this paper, we explore the applicability of Wasserstein Generative Adversarial Networks (WGANs) to SAR data. The latter, being complex, require CV generator and a discriminator taking as input a complex-valued signal, to capture the underlying distribution. In particular, we show the applicability of CV-WGANs for the synthesis of (i) Fourier spectrum of various MNIST like datasets as toy example and (ii) L-band UAVSAR dataset.

Mots clés: Synthetic Aperture Radar Complex Valued Neural Networks Wasserstein Generative Adversarial Networks; Synthetic Aperture Radar; Complex Valued Neural Networks; Wasserstein Generative Adversarial Networks

Domaines: Traitement du signal et de l'image [eess.SP]; Intelligence artificielle [cs.AI]; Electromagnétisme

Victor Dhédin, Jérémie Levi, Jérémy Fix, Chengfang Ren, I. Hinostroza. Complex-valued Wasserstein GAN for SAR Image Generation. IEEE IGARSS 2024, Jul 2024, Athens, Greece. ⟨hal-04529685⟩