Probabilistic Emissivity Retrieval from Hyperspectral Data via Physics-Guided Variational Inference

2025-08-05

AI Summary

Traditional hyperspectral target identification methods yield only single-class predictions, rely heavily on predefined material libraries, and lack interpretability and uncertainty quantification. To address these limitations, we propose a physics-conditioned variational inference framework that integrates atmospheric and background priors into an encoder-decoder probabilistic latent variable model. Our method incorporates a physics-guided loss, in-loop data augmentation, and distribution-matching regularization, enabling Monte Carlo sampling to invert the conditional probability distribution of emissivity spectra. Unlike conventional pixel-wise discriminative approaches, our framework eliminates dependence on known material libraries, supports identification of unknown materials, enables distribution-level compositional matching, and ensures physically consistent uncertainty modeling. Experimental results demonstrate substantial improvements in generalization, interpretability, and spectral inversion accuracy.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models
• Computer Vision: Low Level & Physics-based Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization

Abstract

Recent research has proven neural networks to be a powerful tool for performing hyperspectral imaging (HSI) target identification. However, many deep learning frameworks deliver a single material class prediction and operate on a per-pixel basis; such approaches are limited in their interpretability and restricted to predicting materials that are accessible in available training libraries. In this work, we present an inverse modeling approach in the form of a physics-conditioned generative model.A probabilistic latent-variable model learns the underlying distribution of HSI radiance measurements and produces the conditional distribution of the emissivity spectrum. Moreover, estimates of the HSI scene's atmosphere and background are used as a physically relevant conditioning mechanism to contextualize a given radiance measurement during the encoding and decoding processes. Furthermore, we employ an in-the-loop augmentation scheme and physics-based loss criteria to avoid bias towards a predefined training material set and to encourage the model to learn physically consistent inverse mappings. Monte-Carlo sampling of the model's conditioned posterior delivers a sought emissivity distribution and allows for interpretable uncertainty quantification. Moreover, a distribution-based material matching scheme is presented to return a set of likely material matches for an inferred emissivity distribution. Hence, we present a strategy to incorporate contextual information about a given HSI scene, capture the possible variation of underlying material spectra, and provide interpretable probability measures of a candidate material accounting for given remotely-sensed radiance measurement.

Problem

• Retrieving probabilistic emissivity from hyperspectral data
• Overcoming limitations of single-pixel deep learning predictions
• Incorporating atmospheric and background context in inverse modeling

Innovation

• Physics-guided variational inference for emissivity retrieval
• Conditional generative model with atmospheric context
• Monte-Carlo sampling for uncertainty quantification