Radargrams as Sequences: A Method for The Semantic Segmentation of Radar Sounder Data

Radar sounders (RS) are electromagnetic sensors employed for subsurface investigation on Earth and other celestial bod- ies. The data acquired by RS are represented as radargrams and they are traditionally analyzed through visual interpreta- tion or manual feature engineering to detect subsurface layers and relevant features. The emergence of deep learning has prompted exploration into deep automated radargram anal- ysis, frequently treating them as conventional images. Here we present a novel method for the semantic segmentation of radargrams inspired to video object segmentation (VOS). It consists in an initial phase of self-supervised VOS-based learning of features from sequences of radargram patches, and a second phase of label propagation from an initial reference patch provided by experts. The effectiveness of the proposed method is confirmed by validation on a dataset of radar- grams generated by the airborne radar sounder MCoRDS1 in Antarctica.