Frequency and Edge-Guided Segment Anything Model for Remote Sensing Image Semantic Segmentation

2026-08-15

AI Summary

This study addresses the challenges of insufficient feature adaptation and ambiguous boundaries when applying the Segment Anything Model (SAM) to remote sensing semantic segmentation. We propose FE-SAM, a novel framework that incorporates a frequency-domain modulation adapter to enhance land-cover feature representation and integrates a multi-scale edge refinement module to improve boundary precision, thereby enabling efficient SAM fine-tuning. Extensive experiments demonstrate that FE-SAM significantly outperforms state-of-the-art methods across three mainstream benchmark datasets. By effectively resolving feature alignment and fine-grained segmentation difficulties in remote sensing scenarios, this work establishes a new paradigm for adapting large-scale pre-trained models to the remote sensing domain.

Abstract

Remote sensing image semantic segmentation (RSISS) has attracted significant attention due to the growing demand for fine-grained land cover information. The Segment Anything Model (SAM), proposed as a foundation vision model, offers strong segmentation performance and generalization capabilities for RSISS tasks. However, existing SAM-based approaches face two limitations: (1) Insufficient adaptation of SAM's features to the diverse characteristics of land cover types. (2) Semantic ambiguity at object boundaries, which hinders accurate delineation. To address these limitations, we propose Frequency and Edge-guided SAM (FE-SAM), a scalable and efficient framework for RSISS. Specifically, we introduce a Frequency-Modulated Adapter (FMA) that adaptively decomposes and modulates frequency-domain features based on the input data. It selectively enhances informative high- and low-frequency components corresponding to different land cover types. Furthermore, to improve SAM's ability to capture fine-grained details, we design EGRefiner, which integrates multi-scale edge-enhanced information extracted from the input image. Extensive experiments on three benchmark datasets demonstrate that FE-SAM outperforms state-of-the-art methods. The source codes are available at: https://github.com/oucailab/FE-SAM.

Problem

• Remote Sensing Image Semantic Segmentation
• Segment Anything Model
• Feature Adaptation
• Boundary Ambiguity

Innovation

• Frequency-Modulated Adapter
• EGRefiner
• Segment Anything Model
• Remote Sensing Image Semantic Segmentation
• Edge-Guided