English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Multi-satellite, multi-modality in-orbit fusion is a challenging task as itexplores the fusion representation of complex high-dimensional data underlimited computational resources. Deep neural networks can reveal the underlyingdistribution of multi-modal remote sensing data, but the in-orbit fusion ofmultimodal data is more difficult because of the limitations of differentsensor imaging characteristics, especially when the multimodal data followsnon-independent identically distribution (Non-IID) distributions. To addressthis problem while maintaining classification performance, this paper proposesa manifold-driven multi-modality fusion framework, FedFusion, which randomlysamples local data on each client to jointly estimate the prominent manifoldstructure of shallow features of each client and explicitly compresses thefeature matrices into a low-rank subspace through cascading and additiveapproaches, which is used as the feature input of the subsequent classifier.Considering the physical space limitations of the satellite constellation, wedeveloped a multimodal federated learning module designed specifically formanifold data in a deep latent space. This module achieves iterative updatingof the sub-network parameters of each client through global weighted averaging,constructing a framework that can represent compact representations of eachclient. The proposed framework surpasses existing methods in terms ofperformance on three multimodal datasets, achieving a classification averageaccuracy of 94.35$\%$ while compressing communication costs by a factor of 4.Furthermore, extensive numerical evaluations of real-world satellite imageswere conducted on the orbiting edge computing architecture based on Jetson TX2industrial modules, which demonstrated that FedFusion significantly reducedtraining time by 48.4 minutes (15.18%) while optimizing accuracy.}

FedFusion: Manifold Driven Federated Learning for Multi-satellite and Multi-modality Fusion