Exploring Federated Learning for Thermal Urban Feature Segmentation -- A Comparison of Centralized and Decentralized Approaches

2025-10-28

AI Summary

In urban drone thermal imaging semantic segmentation, cross-regional data cannot be centrally shared due to privacy constraints and technical limitations, exhibiting severe non-IID characteristics. Method: This work presents the first systematic evaluation of federated learning (FL) for thermal image segmentation under realistic deployment conditions. We propose a distributed training framework tailored to thermal imaging properties, integrating model aggregation optimization and communication efficiency enhancements, while comparatively analyzing client- versus server-controlled paradigms. Results: FL achieves segmentation accuracy nearly matching centralized training (mIoU degradation <3.2%) while preserving data locality. We quantitatively characterize communication overhead, training latency, and edge-device energy consumption, revealing significant performance disparities among mainstream FL algorithms under remote-sensing thermal imaging non-IID settings. This study establishes a reproducible methodology and empirical benchmark for privacy-sensitive edge-intelligent remote sensing applications.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Computer Vision: Segmentation
• Search and Optimization: Distributed Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Federated Learning (FL) is an approach for training a shared Machine Learning (ML) model with distributed training data and multiple participants. FL allows bypassing limitations of the traditional Centralized Machine Learning CL if data cannot be shared or stored centrally due to privacy or technical restrictions -- the participants train the model locally with their training data and do not need to share it among the other participants. This paper investigates the practical implementation and effectiveness of FL in a real-world scenario, specifically focusing on unmanned aerial vehicle (UAV)-based thermal images for common thermal feature detection in urban environments. The distributed nature of the data arises naturally and makes it suitable for FL applications, as images captured in two German cities are available. This application presents unique challenges due to non-identical distribution and feature characteristics of data captured at both locations. The study makes several key contributions by evaluating FL algorithms in real deployment scenarios rather than simulation. We compare several FL approaches with a centralized learning baseline across key performance metrics such as model accuracy, training time, communication overhead, and energy usage. This paper also explores various FL workflows, comparing client-controlled workflows and server-controlled workflows. The findings of this work serve as a valuable reference for understanding the practical application and limitations of the FL methods in segmentation tasks in UAV-based imaging.

Problem

• Comparing centralized and federated learning for thermal image segmentation
• Evaluating federated learning performance in real UAV deployment scenarios
• Analyzing federated workflows for urban thermal feature detection

Innovation

• Federated Learning trains models with distributed data
• Compares centralized and decentralized learning approaches
• Evaluates workflows and performance in real scenarios