Spatially-adaptive dynamic dictionary learning for fine-grained remote sensing image segmentation
- 1HuanTian Wisdom Technology Co., Ltd.
Abstract
Dynamic dictionary learning provides explicit class-level semantic prototypes for remote sensing image segmentation. However, most existing methods generate one global dictionary for the whole image through global feature aggregation, which overlooks the fact that the same class may have different local appearances under shadows, scale changes, seasonal variation, or background clutter. To address this limitation, we propose Spatially-Adaptive Dynamic Dictionary Learning (SA-DDL). Instead of following the conventional one-image-one-dictionary design, SA-DDL constructs a dense Dictionary Field in which each local region is assigned its own class embeddings. The framework contains a Local Context Modulator for topology-preserving dictionary generation and a Region-Aware Interaction Decoder for aligning pixel features with their corresponding local dictionary atoms. Experiments on LoveDA, UAVid, Potsdam, and Vaihingen show that SA-DDL consistently improves segmentation accuracy over strong recent baselines, while additional ablation, parameter sensitivity, and perturbation studies verify the effectiveness and robustness of the proposed spatial adaptation mechanism.
Related
- UAV image segmentation in complex mining areas by integrating spatially adaptive feature modulation and multi-scale context fusion modules
- BCFL-DAI-PUB: Blockchain-integrated federated learning with dynamic adaptive incentives for privacy-utility balanced credit risk modelling
- Boundary-interior ratio: a spatially decomposed evaluation framework for CNN and mamba-based breast ultrasound segmentation
- AMF-U-Net: an adaptive multimodal fusion residual attention 3D U-Net for boundary-aware brain tumour segmentation
- AdaSwitch3D: Reliability-Aware Human-in-the-Loop Artificial Intelligence for Breast Tumor Segmentation in Dynamic Contrast-Enhanced MRI Across Unseen Centers
BibTeX
@article{zhang2026spatiallyadaptive,
title = {{Spatially-adaptive dynamic dictionary learning for fine-grained remote sensing image segmentation}},
author = {Yan Zhang and Zifei Luo and Bo Yang and Zhangkai Chen and Yunkun Zou},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-63993-z},
url = {https://doi.org/10.1038/s41598-026-63993-z},
note = {CC BY},
}
RIS
TY - JOUR
TI - Spatially-adaptive dynamic dictionary learning for fine-grained remote sensing image segmentation
AU - Yan Zhang
AU - Zifei Luo
AU - Bo Yang
AU - Zhangkai Chen
AU - Yunkun Zou
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-63993-z
UR - https://doi.org/10.1038/s41598-026-63993-z
AB - Dynamic dictionary learning provides explicit class-level semantic prototypes for remote sensing image segmentation. However, most existing methods generate one global dictionary for the whole image through global feature aggregation, which overlooks the fact that the same class may have different local appearances under shadows, scale changes, seasonal variation, or background clutter. To address this limitation, we propose Spatially-Adaptive Dynamic Dictionary Learning (SA-DDL). Instead of following the conventional one-image-one-dictionary design, SA-DDL constructs a dense Dictionary Field in which each local region is assigned its own class embeddings. The framework contains a Local Context Modulator for topology-preserving dictionary generation and a Region-Aware Interaction Decoder for aligning pixel features with their corresponding local dictionary atoms. Experiments on LoveDA, UAVid, Potsdam, and Vaihingen show that SA-DDL consistently improves segmentation accuracy over strong recent baselines, while additional ablation, parameter sensitivity, and perturbation studies verify the effectiveness and robustness of the proposed spatial adaptation mechanism.
SN - 2045-2322
ER -