2026· IEEE Transactions on Geoscience and Remote Sensing· Vol 64, pp. 5532418-5532418· 0 citations· 57 references
TL;DR
A hyperspectral conditional diffusion for dense (HCD2) classification framework is proposed in this article, which reformulates HSI classification as a conditional label-map denoising process guided by hyperspectral features, enabling generative dense semantic inference.
Abstract
Hyperspectral image (HSI) classification aims to recover accurate land-cover semantic distributions from high-dimensional spectral–spatial observations remaining a critical pillar in intelligent Earth observation. Existing models, such as the convolutional neural networks (CNNs), the Transformers, and the Mamba are predominantly based on discriminative learning paradigms, which often struggle to effectively model the underlying spectral–spatial distributions of ground objects under complex noise, spectral mixing, and intraclass spectral variability. To address the limitations, a hyperspectral conditional diffusion for dense (HCD2) classification framework is proposed in this article, which reformulates HSI classification as a conditional label-map denoising process guided by hyperspectral features, enabling generative dense semantic inference. Specifically, in the encoding stage, an adaptive spectral decoupling (ASD) module is designed to enhance class-discriminative spectral representations. Furthermore, a spatial–spectral joint topological encoder (SSJTE) is constructed to jointly model long-range spatial dependencies and spectral topological structures, thereby learning robust spectral–spatial representations. In the decoding stage, a conditional diffusion category decoder (CDCD) is proposed to achieve semantic category restoration and progressive refinement through conditional diffusion and iterative denoising. Experiments on several public datasets demonstrate that the proposed HCD2 consistently outperforms existing discriminative and generative methods, while exhibiting robustness in complex scenarios.
This work proposes a four–stage hybrid framework for hyperspectral image classification that integrates spectral modeling and spatial feature extraction within a unified pipeline and compares the model with recent baselines, including DaViT.
SMGNet is proposed, a lightweight spectral-Mamba-guided network that establishes a “spectral leads, spatial follows” paradigm that employs a state space model (Mamba) to efficiently capture global long-range spectral dependencies with linear complexity, while a lightweight Res-CNN preserves intact spatial topology.
Xiao-Yang Meng, Long-Bao Wang, Kai-Jun Jia et al.· IEEE Journal of Selected Top...· 0 citations
Hyperspectral image classification (HSIC) aims to discriminate land cover categories accurately by exploiting rich spatial and spectral information. However, existing methods still suffer from insufficient collaborative modeling of spatial and spectral features, as well as inadequate fusion of local and global informat...
Hyperspectral image classification requires a compact representation that can jointly capture discriminative spectral cues, local spatial structures, and efficient inference behavior under limited annotations. Existing CNN-based methods provide effective local pattern extraction but are less capable of modeling long-ra...
Xiao-Fei Yang, Rui-Ying Zhang, Wei-Jia Cao et al.· IEEE Transactions on Geoscie...· 0 citations
A novel knowledge transfer paradigm from classification tasks to anomaly detection is proposed, which employs a pretrained convolutional enhanced Transformer (CET) classification network as the teacher model, transferring its rich spectral–spatial context encodings via feature-level distillation.
Lei Xia, Yin-Nian Liu· Remote Sensing· 0 citations
Given the inherent tradeoff between spectral and spatial resolution, hyperspectral images (HSIs) typically exhibit insufficient spatial details. Combining the HSI with a corresponding high spatial resolution conventional imagery serves as a compromised alternative to produce a high-quality HSI. While fusion-based HSI s...
Fei Ye, Peng Zheng, Yang Xu et al.· IEEE Transactions on Neural...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.