Spectral Property-Driven Data Augmentation for Hyperspectral Single-Source Domain Generalization

Taiqin Chen; Yifeng Wang; Xiaochen Feng; Zhilin Zhu; Hao Sha; Yingjian Li; Yongbing Zhang

cs.CVMar 17, 2026

Spectral Property-Driven Data Augmentation for Hyperspectral Single-Source Domain Generalization

Authors: Taiqin Chen, Yifeng Wang, Xiaochen Feng, Zhilin Zhu, Hao Sha, Yingjian Li, Yongbing Zhang

Organizations: School of Computer Science and Technology, Harbin Institute of Technology (Shenzhen), Shenzhen, China · Pengcheng Laboratory, Shenzhen, China · Department of Computer Science and Technology, Tsinghua University, Beijing, China

Abstract

While hyperspectral images (HSI) benefit from numerous spectral channels that provide rich information for classification, the increased dimensionality and sensor variability make them more sensitive to distributional discrepancies across domains, which in turn can affect classification performance. To tackle this issue, hyperspectral single-source domain generalization (SDG) typically employs data augmentation to simulate potential domain shifts and enhance model robustness under the condition of single-source domain training data availability. However, blind augmentation may produce samples misaligned with real-world scenarios, while excessive emphasis on realism can suppress diversity, highlighting a tradeoff between realism and diversity that limits generalization to target domains. To address this challenge, we propose a spectral property-driven data augmentation (SPDDA) that explicitly accounts for the inherent properties of HSI, namely the device-dependent variation in the number of spectral channels and the mixing of adjacent channels. Specifically, SPDDA employs a spectral diversity module that resamples data from the source domain along the spectral dimension to generate samples with varying spectral channels, and constructs a channel-wise adaptive spectral mixer by modeling inter-channel similarity, thereby avoiding fixed augmentation patterns. To further enhance the realism of the augmented samples, we propose a spatial-spectral co-optimization mechanism, which jointly optimizes a spatial fidelity constraint and a spectral continuity self-constraint. Moreover, the weight of the spectral self-constraint is adaptively adjusted based on the spatial counterpart, thus preventing over-smoothing in the spectral dimension and preserving spatial structure. Extensive experiments conducted on three remote sensing benchmarks demonstrate that SPDDA outperforms state-of-the-art methods.

Figures & tables

Figure 1: Overview of spectral property-driven data augmentation (SPDDA). Briefly, a spectral diversity module (SDM) is utilized to inject perturbations into source-domain samples $X_{\mathrm{SD}}\in\mathbb{R}^{H\times W\times C}$ to obtain extended-domain samples $X_{\mathrm{ED}}\in\mathbb{R}^{H\times W\times K}$ , where $K$ is a random integer. SDM constructs a channel mask and a channel-wise adaptive spectral mixer to masks an arbitrary number of channels and mixes the remaining channels, which adheres to the practical phenomenon in hyperspectral image. The augmentation procedure is supervised by a spatial-spectral co-optimization mechanism, which is comprised of a spatial fidelity constraint and a spectral continuity self-constraint.

Figure 2: Illustration of the variation in the spectral continuity self-constraint weight $\lambda$ .

Methods	Houston			Pavia			HyRank
Methods	OA $\uparrow$	F1 $\uparrow$	Kappa $\uparrow$	OA $\uparrow$	F1 $\uparrow$	Kappa $\uparrow$	OA $\uparrow$	F1 $\uparrow$	Kappa $\uparrow$
SDENet	67.02	0.6768	0.4702	79.05	0.7860	0.7471	58.40	0.5574	0.4916
FDGNet	75.44	0.7329	0.5360	78.28	0.7736	0.7371	56.81	0.5035	0.4554
S2AMSNet	71.13	0.7291	0.5541	73.42	0.7303	0.6792	59.46	0.5561	0.4940
S2ECNet	72.84	0.7355	0.5641	78.05	0.7828	0.7358	60.21	0.5617	0.5047
ProRandConv	75.71	0.7635	0.5911	78.27	0.7763	0.7375	57.76	0.5295	0.4753

Table 1: Quantitative Results on Cross-scene Hyperspectral Image Classification for Remote Sensing. The optimal results are bold, and the sub-optimal results are underlined.

Figure 3: Illustration of the pseudo-color images for the extended-domain scene for each method. Regions marked by the blue rectangle are enlarged and displayed in the lower right corner of each image.

Figure 4: Comparison of realism and diversity. PSNR measures spatial realism, while the mean and standard deviation (Std) of SAM represent spectral realism and diversity, respectively. A lower mean SAM indicates higher spectral realism, and a greater Std means higher diversity.

CASM	$\mathcal{L}_{\mathrm{SF}}$	$\mathcal{L}_{\mathrm{SC}}$	OA $\uparrow$	F1 $\uparrow$	Kappa $\uparrow$
$\times$	$\times$	$\times$	74.73	0.7637	0.6068
$\checkmark$	$\times$	$\times$	76.95	0.7794	0.6264
$\times$	$\checkmark$	$\times$	70.93	0.7252	0.5408
$\times$	$\times$	$\checkmark$	72.23	0.7414	0.5711
$\checkmark$	$\checkmark$	$\times$	77.87	0.7724	0.6081
$\checkmark$	$\times$	$\checkmark$	75.94	0.7643	0.6003

Table 2: Ablation study of channel-wise adaptive spectral mixer (CASM), spatial fidelity constraint (

\mathcal{L}_{\mathrm{SF}}

), and spectral continuity self-constraint (

\mathcal{L}_{\mathrm{SC}}

) on Houston.

Figure 5: Effects of CASM, $\mathcal{L}_{\mathrm{SF}}$ , and $\mathcal{L}_{\mathrm{SC}}$ on spectral angle mapper (SAM). The average SAM quantifies realism of generation. A lower average SAM indicates higher realism.

$\lambda$ -T	OA $\uparrow$	F1 $\uparrow$	Kappa $\uparrow$
Fix-0.1	75.19	0.7563	0.5677
Fix-0.5	75.18	0.7651	0.5877
Fix-1.0	72.05	0.7398	0.5374
Fix-1.5	75.27	0.7662	0.5800
Fix-2.0	74.47	0.7620	0.5729
$\mathrm{Var}_{\mathrm{E}}$	76.15	0.7632	0.5906

Table 3: Ablation study of

\lambda

-tuning (

\lambda

-T) strategy on Houston. Fix-0.1 denotes that

\lambda

is set to a fixed value 0.1.

\mathrm{Var}_{\mathrm{E}}

and

\mathrm{Var}_{\mathrm{S}}

represent early–late stage division based on training epochs and

\|\mathcal{L}_{\mathrm{SF}}\|_{1}

, respectively, with

\lambda

set to 0.1 in the early stage and 2.0 in the late stage.

Explore similar work

May 13, 2026

HIR-ALIGN: Enhancing Hyperspectral Image Restoration via Diffusion-Based Data Generation

Hyperspectral image (HSI) restoration is crucial for reliable analysis, as real-world HSIs suffer from noise, blur, and resolution loss. However, existing models trained on source data often fail on target domains lacking clean references, a common real-world scenario. To address this, we present HIR-ALIGN, a plug-and-play target-adaptive augmentation framework that enhances HSI restoration by augmenting limited training images with synthetic data matching the target distribution, without extra clean target-domain HSI data. It has three stages: (i) proxy generation, where off-the-shelf restoration models are applied to degraded target observations to produce semantics-preserving proxy HSIs that approximate clean target-domain images; (ii) distribution-adaptive synthesis, where a blur-robust unCLIP diffusion model generates target-aligned RGBs from proxy RGBs with prompt conditioning and embedding-space noise initialization. The warp-based spectral transfer module then synthesizes HSIs by aligning each generated RGB with its proxy RGB, estimating soft patch-wise transport weights, and applying these weights and learnable local interpolation kernels to the proxy HSI; and (iii) aligned supervised finetuning, where restoration networks pretrained on the source distribution are finetuned with proxy HSIs and synthesized target-aligned HSIs, then deployed on degraded target images. We also provide theoretical analysis showing that, under stated assumptions, the proposed augmentation-based finetuning obtains a tighter target-domain restoration-risk upper bound by jointly improving target-distribution coverage and controlling spectral bias. Experiments on simulated and real datasets across denoising, super-resolution, and other restoration tasks demonstrate that HIR-ALIGN is superior to proxy-only target-adaptation baselines and outperforms representative unsupervised methods in most cases.

Li Pang, Heng Zhao, Yijia Zhang +2

Synthetic Data Augmentation Unsupervised Domain Adaptation Hyperspectral Imaging

Ask PDF

Apr 29, 2026

High-Dimensional Noise to Low-Dimensional Manifolds: A Manifold-Space Diffusion Framework for Degraded Hyperspectral Image Classification

Recently, Hyperspectral Image (HSI) classification has attracted increasing attention in remote sensing. However, HSI data are inherently high-dimensional but low-rank, with discriminative information concentrated on a low-dimensional latent manifold. In real-world remote sensing scenarios, the superposition of multiple degradation factors disrupts this intrinsic manifold structure, driving samples away from their original low-dimensional distribution and introducing substantial redundant and non-discriminative variations. To better handle this challenge, this paper proposes a manifold-space diffusion framework (MSDiff) for robust hyperspectral classification under complex degradation conditions. Specifically, the proposed method first maps high-dimensional, degradation-affected HSI data into a compact low-dimensional manifold through a discriminative spectral-spatial reconstruction task, preserving class semantics and reducing redundant variations. A diffusion-based generative model is then applied to regularize the spectral-spatial distribution within the manifold, enabling progressive refinement and stabilization of latent features against residual degradations. The key advantage of the proposed framework lies in performing diffusion-based distribution modeling directly on the low-dimensional manifold, effectively decoupling degradation-induced disturbances from intrinsic discriminative structures and enhancing representation stability under complex degradations. Experimental results on multiple hyperspectral benchmarks demonstrate consistent performance improvements over state-of-the-art methods under diverse composite degradation settings. The code will be available at https://github.com/yangboxiang1207/MSDiff

Boxiang Yang, Ning Chen, Xia Yue +6

Institute of Remote Sensing and Geographic Information System, Peking University, Beijing 100871, China · Beijing Key Laboratory of Spatio-temporal Perception and Urban Resilience, Peking University, Beijing 100871, China · School of Computer Science and Engineering, Central South University, Changsha 410083, China +4

Manifold Learning Hyperspectral Image Classification Diffusion Models

Ask PDF

May 5, 2026

SoDa2: Single-Stage Open-Set Domain Adaptation via Decoupled Alignment for Cross-Scene Hyperspectral Image Classification

Cross-scene hyperspectral image (HSI) classification stands as a fundamental research topic in remote sensing, with extensive applications spanning various fields. Owing to the inclusion of unknown categories in the target domain and the existence of domain shift across different scenes, open-set domain adaptation techniques are commonly employed to address cross-scene HSI classification. However, existing open-set cross-scene HSI classification methods still face two critical challenges: (1) domain shift issues arising from the direct alignment of mixed spectral-spatial features; (2) high computational costs caused by two-stage training strategies. To address these issues, this paper proposes a single-stage open-set domain adaptation method with decoupled alignment (SoDa

^2

) for cross-scene HSI classification. A contribution-aware dual-modality feature extraction is customized to disentangle the characteristics from spectral sequence signals and spatial details, selectively and adaptively enhancing discriminative features. The decoupled alignment module minimizes the Maximum Mean Discrepancy to independently reduce the spectral discrepancy and the spatial discrepancy between the source and target domains, extracting more fine-grained domain-invariant features. A cost-effective single-stage dual-branch framework is designed to learn MMD-constrainted aligned features and constraint-free intrinsic features for adaptive distinction between known and unknown classes. This framework employs a Gaussian Mixture Model to model the squared cosine similarity distribution between the two feature types, enabling open-set recognition without prior knowledge of unknown classes. Extensive experiments on three groups of HSI datasets demonstrate that SoDa

^2

outperforms state-of-the-art methods, achieving superior classification accuracy and model transferability for open-set cross-scene tasks.

Yiwen Liu, Minghua Wang, Jing Yao +2

Institute of Robotics and Automatic Information System (IRAIS), College of Artificial Intelligence, and the Tianjin Key Laboratory of Intelligent Robotics (tjKLIR), Nankai University, Tianjin 300071, China · Aerospace Information Research Institute, Chinese Academy of Sciences, 100094 Beijing, China · Institute of Methodologies for Environmental Analysis (CNR-IMAA), National Research Council, 85050 Tito, Italy

Unsupervised Domain Adaptation Domain Adaptation Hyperspectral Image Classification

Ask PDF