IGraSS: Learning to Identify Infrastructure Networks from Satellite Imagery by Iterative Graph-constrained Semantic Segmentation

2025-06-09

AI Summary

To address topological distortion in infrastructure network segmentation (e.g., irrigation canals, roads) from satellite imagery caused by label scarcity or noise, this paper proposes a graph-constrained iterative self-refinement semantic segmentation framework. Methodologically, it fuses multi-modal features—RGB, NDWI, and DEM—to build a segmentation model; introduces a global graph-theoretic constraint module leveraging minimum spanning trees and connected components to dynamically refine pseudo-labels using reachability and connectivity properties; and jointly optimizes segmentation predictions and graph structure through iterative co-refinement under weak supervision. In irrigation canal extraction, the proportion of unreachable segments drops significantly from 18% to 3%. Cross-domain validation on road completion further demonstrates strong generalizability. This work is the first to embed graph-structural priors into an iterative weakly supervised segmentation pipeline, substantially improving geometric integrity and robustness in remote sensing infrastructure recognition.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Graph-based Machine Learning
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Accurate canal network mapping is essential for water management, including irrigation planning and infrastructure maintenance. State-of-the-art semantic segmentation models for infrastructure mapping, such as roads, rely on large, well-annotated remote sensing datasets. However, incomplete or inadequate ground truth can hinder these learning approaches. Many infrastructure networks have graph-level properties such as reachability to a source (like canals) or connectivity (roads) that can be leveraged to improve these existing ground truth. This paper develops a novel iterative framework IGraSS, combining a semantic segmentation module-incorporating RGB and additional modalities (NDWI, DEM)-with a graph-based ground-truth refinement module. The segmentation module processes satellite imagery patches, while the refinement module operates on the entire data viewing the infrastructure network as a graph. Experiments show that IGraSS reduces unreachable canal segments from around 18% to 3%, and training with refined ground truth significantly improves canal identification. IGraSS serves as a robust framework for both refining noisy ground truth and mapping canal networks from remote sensing imagery. We also demonstrate the effectiveness and generalizability of IGraSS using road networks as an example, applying a different graph-theoretic constraint to complete road networks.

Problem

• Accurate mapping of canal networks for water management
• Overcoming incomplete ground truth in infrastructure mapping
• Leveraging graph properties to refine segmentation and ground truth

Innovation

• Iterative graph-constrained semantic segmentation framework
• Combines RGB, NDWI, and DEM modalities
• Graph-based ground-truth refinement module