Federated Semantic Segmentation of Synthetic UAV Imagery under Non-IID Conditions
DOI:
https://doi.org/10.22456/2175-2745.150921Keywords:
federated learning, semantic segmentation, UAV control, non-IID dataAbstract
Unmanned Aerial Vehicles (UAVs) rely on semantic segmentation for monitoring, inspection, and disaster response tasks. Traditional centralized training requires the transfer of large data volumes, which is often infeasible because of privacy, bandwidth, and decentralization constraints. Federated Learning (FL) offers a promising alternative; however, non-IID data hinder its effectiveness because UAVs operate in diverse environments with heterogeneous distributions. This study evaluated the adaptation of a semantic segmentation model to a federated setting using synthetic UAV imagery. We investigated four training configurations by varying the communication rounds, local epochs, and learning rates, and analyzed their effects on convergence and class-level accuracy. The results suggest that moderate communication with local training achieves balanced performance, whereas high learning rates or excessive synchronization cause instability and class-specific biases. These findings open avenues for designing robust federated UAV segmentation strategies under non-IID conditions.
Downloads
References
[1] BERGMANN, M. A. et al. An approach based on lidar and spherical images for automated vegetation inspection in urban power distribution lines. IEEE Access, v. 12, p. 105119–105130, 2024.
[2] SILVA, L. et al. Photo-realistic and labeled synthetic uav flight data generation using ros and gazebo. In: 2024 Latin American Robotics Symposium (LARS). Arequipa, Peru: IEEE, 2024. p. 1–6.
[3] EPIFANI, L.; CARUSO, A. A survey on deep learning in uav imagery for precision agriculture and wild flora monitoring: Datasets, models and challenges. Smart Agricultural Technology, v. 9, p. 100625, 2024. ISSN 2772-3755. Disponível em: https://www.sciencedirect.com/science/article/pii/S2772375524002302.
[4] THAKUR, D. et al. Green federated learning: A new era of green aware ai. ACM Comput. Surv., Association for Computing Machinery, New York, NY, USA, v. 57, n. 8, mar. 2025. ISSN 0360-0300. Disponível em: https://doi.org/10.1145/3718363.
[5] MCMAHAN, B. et al. Communication-Efficient Learning of Deep Networks from Decentralized Data. In: SINGH, A.; ZHU, J. (Ed.). Proceedings of the 20th International Conference on Artificial Intelligence and Statistics. PMLR, 2017. (Proceedings of Machine Learning Research, v. 54), p. 1273–1282. Disponível em: https://proceedings.mlr.press/v54/mcmahan17a.html.
[6] ANTICO, T.; RODRIGUES MOREIRA, L. F.; MOREIRA, R. Evaluating the Potential of Federated Learning for Maize Leaf Disease Prediction. In: Anais do XIX Encontro Nacional de Inteligência Artificial e Computacional. Porto Alegre, RS, Brasil: SBC, 2022. p. 282–293. ISSN 2763-9061. Disponível em: https://sol.sbc.org.br/index.php/eniac/article/view/22789.
[7] RODRIGUES, L. G. F. et al. Medical image classification with privacy: Centralized and federated learning comparison. Revista de Informática Teórica e Aplicada, v. 32, n. 1, p. 180–187, Feb. 2025. Disponível em: https://seer.ufrgs.br/index.php/rita/article/view/143478.
[8] MOREIRA, R. et al. An intelligent native network slicing security architecture empowered by federated learning. Future Generation Computer Systems, v. 163, p. 107537, 2025. ISSN 0167-739X. Disponível em: https://www.sciencedirect.com/science/article/pii/S0167739X24005016.
[9] BARBOSA, G. V. G. et al. Federated learning in breast cancer diagnosis. Revista de Informática Teórica e Aplicada, v. 32, n. 1, p. 173–179, 2025. Disponível em: https://seer.ufrgs.br/index.php/rita/article/view/143480.
[10] UDDIN, M. P. et al. A systematic literature review of robust federated learning: Issues, solutions, and future research directions. ACM Comput. Surv., Association for Computing Machinery, New York, NY, USA, v. 57, n. 10, maio 2025. ISSN 0360-0300. Disponível em: https://doi.org/10.1145/3727643.
[11] QUIGLEY, M. et al. ROS: an open-source Robot Operating System. In: ICRA Workshop on Open Source Software. Kobe, Japan: IEEE, 2009. v. 3, n. 3.2, p. 5.
[12] KOENIG, N.; HOWARD, A. Design and use paradigms for gazebo, an open-source multi-robot simulator. In: 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566). Sendai, Japan: IEEE, 2004. v. 3, p. 2149–2154 vol. 3.
[13] ULLAH, W. et al. Split federated learning for real-time aerial video event recognition in uav-based geospatial monitoring. IEEE Transactions on Geoscience and Remote Sensing, p. 1–1, 2025. ISSN 1558-0644.
[14] MIAO, J. et al. Fedseg: Class-heterogeneous federated learning for semantic segmentation. In: 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Vancouver, BC, Canada: IEEE, 2023. p. 8042–8052.
[15] ALSHAMMARI, G. K. et al. Federated Learning-based Semantic Segmentation for Lane and Object Detection in Autonomous Driving. 2025. Disponível em: https://arxiv.org/abs/2504.18939.
[16] IMANDI, R. et al. Building sustainable federated learning models in fog-enabled uav-as-a-service for aerial image classification. Sustainable Computing: Informatics and Systems, v. 46, p. 101133, 2025. ISSN 2210-5379. Disponível em: https://www.sciencedirect.com/science/article/pii/S221053792500054X.
[17] GODAVARTHI, D. et al. Federated learning-based semantic segmentation framework for sustainable development. Egyptian Informatics Journal, v. 30, p. 100702, 2025. ISSN 1110-8665. Disponível em: https://www.sciencedirect.com/science/article/pii/S1110866525000957.
[18] XIE, R. et al. A novel federated learning framework for semantic segmentation of terminal block in smart substation. Pattern Recognition, v. 167, p. 111665, 2025. ISSN 0031-3203. Disponível em: https://www.sciencedirect.com/science/article/pii/S0031320325003255.
[19] WANG, J. et al. Deep high-resolution representation learning for visual recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, v. 43, n. 10, p. 3349–3364, 2021.
[20] BALDIN, I. et al. FABRIC: A national-scale programmable experimental network infrastructure. IEEE Internet Computing, IEEE, v. 23, n. 6, p. 38–47, 2019.
[21] BEUTEL, D. J. et al. Flower: A Friendly Federated Learning Research Framework. 2022. Disponível em: https://arxiv.org/abs/2007.14390.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2026 Victor Gonçalves Vieira, Emanuel Teixeira Martins, Juliana Quintiliano Ferreira, Thiago Luange Gomes, Michel Melo Silva, Rodrigo Moreira, Larissa Ferreira Rodrigues Moreira
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Autorizo aos editores a publicação de meu artigo, caso seja aceito, em meio eletrônico de acordo com as regras do Public Knowledge Project.
