SAFE: a SAR Feature Extractor based on self-supervised learning and masked Siamese ViTs

TL;DR

SAFE is a self-supervised SAR feature extractor using masked Siamese Vision Transformers that learns robust features from unlabeled SAR data, improving performance across multiple tasks without requiring labeled datasets.

Contribution

The paper introduces a novel self-supervised learning framework with SAR-specific data augmentation for generalizable feature extraction using masked Siamese ViTs.

Findings

Outperforms state-of-the-art in few-shot classification

Effective across various SAR modes and resolutions

Enables high-quality segmentation and pattern detection

Abstract

Due to its all-weather and day-and-night capabilities, Synthetic Aperture Radar imagery is essential for various applications such as disaster management, earth monitoring, change detection and target recognition. However, the scarcity of labeled SAR data limits the performance of most deep learning algorithms. To address this issue, we propose a novel self-supervised learning framework based on masked Siamese Vision Transformers to create a General SAR Feature Extractor coined SAFE. Our method leverages contrastive learning principles to train a model on unlabeled SAR data, extracting robust and generalizable features. SAFE is applicable across multiple SAR acquisition modes and resolutions. We introduce tailored data augmentation techniques specific to SAR imagery, such as sub-aperture decomposition and despeckling. Comprehensive evaluations on various downstream tasks, including few-shot classification, segmentation, visualization, and pattern detection, demonstrate the effectiveness and versatility of the proposed approach. Our network competes with or surpasses other state-of-the-art methods in few-shot classification and segmentation tasks, even without being trained on the sensors used for the evaluation.