.png){kind=icon}
.png){kind=icon}
Return
Fractal-Domain Vision Graph Neural Network for Remote Sensing Ground Target Classification
.png){kind=icon}
PRE
AI
J
T
G
W
DOI：10.1109/tpami.2026.3690544.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
To the best of our knowledge, this paper is the first to integrate fractal signal processing with vision graph neural networks, establishing a new graph representation learning paradigm consistent with fractal dynamics. Building on this foundation, we propose a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Fractal-domain Vision Graph Neural Network</i> (FD-ViG). Specifically, FD-ViG includes: (i) a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Fractal-Domain Learning Module</i> that maps images into the fractal-domain using local Hölder exponents and the Singularity Power Spectrum (SPS), enabling fractal–spatial feature fusion; (ii) a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Fractal Graph Construction Module</i> that adaptively generates a topology by combining semantic attention with fractal similarity in the fractal feature space; and (iii) a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Graph Propagation Module</i> with power-law multi-scale propagation to realize cross-scale diffusion and aggregation, enabling coupled texture–structure learning. Experiments on UCMerced, RSSCN7, and SIRI-WHU achieve overall accuracies of 91.75%, 89.52%, and 92.78%, respectively. Compared with representative vision graph models such as ViG, WiGNet, and ViHGNN, our method achieves consistent improvements over prior methods across all three datasets, while remaining lightweight (2.6 M parameters). Moreover, despite having far fewer parameters than ResNet-18, our model yields competitive or better performance on two datasets, and further demonstrates strong generalization ability in cross-dataset evaluation on SAR imagery. This work provides a principled and effective bridge between fractal theory and graph deep learning, benefiting interpretable remote sensing scene understanding under complex textures and structures.
Keywords:
Stochastic fractals
fractal graph neural networks
fractal attention
fractal graph topology
remote sensing
Journal
IF:
18.6
Papers:
831
Citations:
9.8W
Organization
Cited Papers