BASeg: Boundary-Aware Remote Sensing Segmentation with Structural Penalties

2026-08-16

AI Summary

This study addresses the challenges of ambiguous boundaries and limited urban morphological diversity in remote sensing semantic segmentation by proposing the BASeg framework and Mahalanobis Angular Boundary Loss (MABL). The method integrates a global visual state space with cross-feature modules to simultaneously capture long-range dependencies and local details, thereby enhancing boundary shape consistency. Furthermore, we introduce GCD-25k, a benchmark dataset encompassing ten global cities to mitigate data bias. Experiments demonstrate that BASeg achieves up to a 2.8% mIoU improvement across four benchmarks with significantly refined boundary precision. Additionally, MABL exhibits robustness and generalizability across diverse architectures, establishing a novel paradigm for fine-grained remote sensing interpretation.

Abstract

Semantic segmentation is a core computer vision task in the remote sensing field, accelerating advancements in ur- ban development, agriculture, ecology, water resources, and environmental monitoring. However, recent methods usually struggle to capture fine-grained object features and bound- ary details. Besides, current widely used datasets often lack city morphology diversity and segmentation on generative im- ages remains largely unexplored. To address these issues, we propose a Mahalanobis-Angle Boundary Loss (MABL) that explicitly enhances boundary and shape consistency. MABL jointly models structural importance and boundary orientation through Mahalanobis distance-based weighting and angle- aware penalty. It can be readily integrated into diverse seg- mentation architectures and consistently improves their accu- racy. Built upon MABL, we introduce BASeg, a boundary- aware remote sensing segmentation framework with Struc- tural Penalties. BASeg integrates a Global Visual State Space module (GSM) with a Cross-Feature Fusion module (CFM) to capture both long-range contextual dependencies and fine- grained local details. Additionally, we establish a global 10- city benchmark dataset (GCD-25k) to facilitate accurate build- ing and road segmentation. Extensive experiments on four remote-sensing benchmarks demonstrate that BASeg consis- tently outperforms existing methods, achieving up to a 2.8% improvement in mIoU while producing more accurate object boundary segmentation across diverse scenes. Moreover, integrating MABL into multiple existing segmentation archi- tectures consistently improves performance across datasets, demonstrating its robustness and broad applicability.

Problem

• Remote Sensing Segmentation
• Boundary Details
• Fine-grained Features
• Dataset Diversity

Innovation

• Mahalanobis-Angle Boundary Loss
• Boundary-Aware Segmentation
• Visual State Space
• Cross-Feature Fusion
• Remote Sensing Benchmark