Federated learning improves remote sensing image classification under varied conditions
FedMAD: Modulation-Aware Directional Aggregation for Federated Learning in Remote Sensing Image Classification
Computer Vision and Pattern RecognitionMachine Learning
Summary
Remote sensing images come from many places and times, causing differences that make learning models hard to generalize. The authors propose a method called FedMAD that splits model parts into shared global features and local client-specific features to keep unique data patterns. Their approach also smartly adjusts how updates from different clients combine, focusing more on consistent contributions and ignoring conflicting ones. Tests on big remote sensing datasets show this method handles data differences better than other federated learning approaches.
What this means in practice
- •For geospatial data teams: Enable collaborative training of image classifiers across multiple satellite data centers with different environmental conditions without sharing raw data.
- •For environmental monitoring groups: Improve model accuracy in classifying land use and vegetation types across regions by handling geographical and seasonal data differences in decentralized data setups.
Authors
Barış Büyüktaş, Begüm Demir
Abstract
Federated learning (FL) has recently attracted increasing attention in remote sensing (RS) since it enables collaborative model training across decentralized RS image archives without requiring direct access to local data. However, FL performance significantly degrades when the data distributions between clients are heterogeneous, which often occurs due to geographical differences, seasonal changes, and varying image acquisition and atmospheric conditions. To address this challenge, in this letter, we propose a novel personalized FL framework (denoted as FedMAD) for RS image classification problems. The proposed framework separates globally shared representation parameters from client-specific adaptation parameters to preserve client-specific features while maintaining globally transferable representations. This is achieved by integrating lightweight modulation modules and local batch normalization layers into the backbone network. Although globally shared parameters are collaboratively optimized between clients, client-specific parameters remain local to preserve domain-specific feature characteristics. In addition, FedMAD introduces a modulation-aware directional aggregation strategy that dynamically adjusts the importance of aggregation for each client according to the alignment of local modulation updates. This allows the global optimization process to suppress conflicting client updates originating from heterogeneous data distributions while enhancing the contribution of clients with consistent adaptation behaviors. The experimental results obtained on the BigEarthNet-S2 and EuroSAT datasets demonstrate the effectiveness of FedMAD compared to state-of-the-art FL algorithms under heterogeneous RS data distributions. The code of the proposed framework will be publicly available at https://git.tu-berlin.de/rsim/fedmad.