Diffusion Posterior Sampler for Hyperspectral Unmixing with Spectral Variability Modeling

2025-12-10

AI Summary

To address the challenges of modeling spectral variability in endmembers and inaccurate Bayesian posterior estimation of abundances in hyperspectral unmixing, this paper proposes DPS4Un, a semi-blind unmixing method. Methodologically: (1) superpixels are leveraged to construct image-adaptive endmember bundle priors, explicitly encoding spectral variability; (2) a pre-trained conditional spectral diffusion model is introduced for the first time as a Bayesian posterior sampler, generating abundance distributions conditioned on observed data; (3) endmembers—initialized from Gaussian noise—and abundances are jointly optimized within a framework incorporating a data-fidelity term. Evaluated on three real-world hyperspectral benchmark datasets, DPS4Un consistently outperforms state-of-the-art methods, demonstrating the effectiveness of learned spectral priors and diffusion-driven posterior sampling.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Diffusion Models for Vision
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Linear spectral mixture models (LMM) provide a concise form to disentangle the constituent materials (endmembers) and their corresponding proportions (abundance) in a single pixel. The critical challenges are how to model the spectral prior distribution and spectral variability. Prior knowledge and spectral variability can be rigorously modeled under the Bayesian framework, where posterior estimation of Abundance is derived by combining observed data with endmember prior distribution. Considering the key challenges and the advantages of the Bayesian framework, a novel method using a diffusion posterior sampler for semiblind unmixing, denoted as DPS4Un, is proposed to deal with these challenges with the following features: (1) we view the pretrained conditional spectrum diffusion model as a posterior sampler, which can combine the learned endmember prior with observation to get the refined abundance distribution. (2) Instead of using the existing spectral library as prior, which may raise bias, we establish the image-based endmember bundles within superpixels, which are used to train the endmember prior learner with diffusion model. Superpixels make sure the sub-scene is more homogeneous. (3) Instead of using the image-level data consistency constraint, the superpixel-based data fidelity term is proposed. (4) The endmember is initialized as Gaussian noise for each superpixel region, DPS4Un iteratively updates the abundance and endmember, contributing to spectral variability modeling. The experimental results on three real-world benchmark datasets demonstrate that DPS4Un outperforms the state-of-the-art hyperspectral unmixing methods.

Problem

• Develops a diffusion posterior sampler for hyperspectral unmixing
• Models spectral variability using superpixel-based endmember bundles
• Refines abundance distribution by combining learned priors with observations

Innovation

• Diffusion posterior sampler for semiblind unmixing
• Image-based endmember bundles within superpixels
• Superpixel-based data fidelity term for consistency