# Enhanced LULC Segmentation via Lightweight Model Refinements on ALOS-2 SAR Data

arXiv: https://arxiv.org/abs/2601.15705 · HTML: https://mlsanity.com/paper/2601.15705/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 23 references that are in the map.
- First posted on arXiv 22 Jan 2026.

## Builds on (most cited first, 13 of 13)

- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation](https://mlsanity.com/paper/1606.04797/) (2016)
- [Unified Perceptual Parsing for Scene Understanding](https://mlsanity.com/paper/1807.10221/) (2018)
- [Swin Transformer V2: Scaling Up Capacity and Resolution](https://mlsanity.com/paper/2111.09883/) (2021)
- [SpectralGPT: Spectral Remote Sensing Foundation Model](https://mlsanity.com/paper/2311.07113/) (2023)
- [Geography-Aware Self-Supervised Learning](https://mlsanity.com/paper/2011.09980/) (2020)
- [Advancing Plain Vision Transformer Towards Remote Sensing Foundation Model](https://mlsanity.com/paper/2208.03987/) (2022)
- [Foundation Models for Generalist Geospatial Artificial Intelligence](https://mlsanity.com/paper/2310.18660/) (2023)
- [Rethinking Transformers Pre-training for Multi-Spectral Satellite Imagery](https://mlsanity.com/paper/2403.05419/) (2024)
- [MixMAE: Mixed and Masked Autoencoder for Efficient Pretraining of Hierarchical Vision Transformers](https://mlsanity.com/paper/2205.13137/) (2022)
- [Predicting Gradient is Better: Exploring Self-Supervised Learning for SAR ATR with a Joint-Embedding Predictive Architecture](https://mlsanity.com/paper/2311.15153/) (2023)
- [SARFormer -- An Acquisition Parameter Aware Vision Transformer for Synthetic Aperture Radar Data](https://mlsanity.com/paper/2504.08441/) (2025)

## Built on by (most cited first, 1 of 1)

- [A Tutorial on ALOS2 SAR Utilization: Dataset Preparation, Self-Supervised Pretraining, and Semantic Segmentation](https://mlsanity.com/paper/2603.15119/) (2026)

## Abstract (arXiv)

This work focuses on national-scale land-use/land-cover (LULC) semantic segmentation using ALOS-2 single-polarization (HH) SAR data over Japan, together with a companion binary water detection task. Building on SAR-W-MixMAE self-supervised pretraining [1], we address common SAR dense-prediction failure modes, boundary over-smoothing, missed thin/slender structures, and rare-class degradation under long-tailed labels, without increasing pipeline complexity. We introduce three lightweight refinements: (i) injecting high-resolution features into multi-scale decoding, (ii) a progressive refine-up head that alternates convolutional refinement and stepwise upsampling, and (iii) an α-scale factor that tempers class reweighting within a focal+dice objective. The resulting model yields consistent improvements on the Japan-wide ALOS-2 LULC benchmark, particularly for under-represented classes, and improves water detection across standard evaluation metrics.