A Tutorial on ALOS2 SAR Utilization: Dataset Preparation, Self-Supervised Pretraining, and Semantic Segmentation

TL;DR

This paper presents a comprehensive approach to utilizing ALOS-2 SAR data for semantic segmentation, including dataset preparation, self-supervised pretraining with SAR-specific methods, and demonstrating improved performance over traditional training.

Contribution

It introduces SAR-W-SimMIM, a novel self-supervised pretraining method tailored for SAR imagery, and provides a dataset and methodology for fine-tuning models for land cover segmentation.

Findings

Pretraining with SAR-W-SimMIM improves segmentation accuracy.

Constructed a Japan-focused ALOS-2 SAR dataset for pretraining.

Fine-tuned models outperform those trained from scratch.

Abstract

Masked auto-encoders (MAE) and related approaches have shown promise for satellite imagery, but their application to synthetic aperture radar (SAR) remains limited due to challenges in semantic labeling and high noise levels. Building on our prior work with SAR-W-MixMAE, which adds SAR-specific intensity-weighted loss to standard MixMAE for pretraining, we also introduce SAR-W-SimMIM; a weighted variant of SimMIM applied to ALOS-2 single-channel SAR imagery. This method aims to reduce the impact of speckle and extreme intensity values during self-supervised pretraining. We evaluate its effect on semantic segmentation compared to our previous trial with SAR-W-MixMAE and random initialization, observing notable improvements. In addition, pretraining and fine-tuning models on satellite imagery pose unique challenges, particularly when developing region-specific models. Imbalanced land cover distributions such as dominant water, forest, or desert areas can introduce bias, affecting both pretraining and downstream tasks like land cover segmentation. To address this, we constructed a SAR dataset using ALOS-2 single-channel (HH polarization) imagery focused on the Japan region, marking the initial phase toward a national-scale foundation model. This dataset was used to pretrain a vision transformer-based autoencoder, with the resulting encoder fine-tuned for semantic segmentation using a task-specific decoder. Initial results demonstrate significant performance improvements compared to training from scratch with random initialization. In summary, this work provides a guide to process and prepare ALOS2 observations to create dataset so that it can be taken advantage of self-supervised pretraining of models and finetuning downstream tasks such as semantic segmentation.