Privacy-Preserving Federated Land-Cover Segmentation Under Non-IID Data: Region-Aware Contrastive Regularization and DP-SGD
Electronics · 12.11.2026 · 15 (22) · 5241
DOI: 10.3390/electronics15225241 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 12.11.2026. Verfasst von Mustafa Ali Ahmed und 1 weiteren Autor:innen. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Federated learning supports collaborative satellite-image segmentation without centralizing source imagery, but non-IID data distributions and information leakage through model updates remain important challenges. We evaluate centralized U-Net, vanilla FedAvg, client-local region-aware contrastive FedAvg, client-side DP-SGD, and DP-SGD with fixed-prototype contrastive regularization on the DeepGlobe dataset using a fixed five-client Dirichlet partition (α = 0.5). Non-private contrastive FedAvg achieved the highest observed held-out test performance, with mIoU = 0.401 and pixel accuracy = 0.728, compared with mIoU = 0.352 for vanilla FedAvg. DP-SGD provided a formal patch-level privacy guarantee under the stated accounting assumptions but substantially reduced segmentation utility: standalone private training achieved mIoU = 0.138 at ε ≈ 4.82, while the matched private contrastive configuration achieved mIoU = 0.115 at the same privacy expenditure. Private training also incurred substantially greater computational costs. These single-run findings highlight a pronounced privacy–utility–computation trade-off and warrant validation across multiple seeds and client partitions.
Zitieren
APA
Ahmed, M. A., & Khader, M. (2026). Privacy-Preserving Federated Land-Cover Segmentation Under Non-IID Data: Region-Aware Contrastive Regularization and DP-SGD. Electronics, 15(22), 5241. https://doi.org/10.3390/electronics15225241
Vancouver
Ahmed MA, Khader M. Privacy-Preserving Federated Land-Cover Segmentation Under Non-IID Data: Region-Aware Contrastive Regularization and DP-SGD. Electronics. 2026;15(22):5241. doi:10.3390/electronics15225241
BibTeX
@article{Ahmed2026,
author = {Ahmed, Mustafa Ali and Khader, Mohammed},
title = {Privacy-Preserving Federated Land-Cover Segmentation Under Non-IID Data: Region-Aware Contrastive Regularization and DP-SGD},
journal = {Electronics},
year = {2026},
volume = {15},
number = {22},
pages = {5241},
doi = {10.3390/electronics15225241}
}
EUVIMED