PNEC-Mamba: Prototype-Guided Positive-Negative Evidence Calibration for Hyperspectral Image Classification

2026-08-03

AI Summary

This work addresses the challenges of hyperspectral image classification, which is often hindered by spectral confusion, mixed pixels, and local interference, leading to ambiguous pixel representations and unreliable decisions. To tackle these issues, the study introduces evidence reliability modeling into this domain for the first time and proposes a prototype-guided positive-negative evidence calibration framework. The method leverages dynamic class prototypes and a pixel-prototype competition mechanism to disentangle discriminative evidence from interfering factors, while incorporating multi-source uncertainty estimation for selective evidence calibration. Furthermore, full-resolution consistency constraints are employed to refine boundary details. Evaluated on three benchmark datasets, the proposed approach significantly outperforms current state-of-the-art models, achieving notable improvements in both classification accuracy and spatial boundary coherence.

Abstract

In real-world hyperspectral scenes, pixel representations are often ambiguous due to factors such as spectral similarity, mixed pixels, and local context interference, which may simultaneously encode discriminative evidence and interfering information. Existing methods mainly focus on learning more powerful representations or modeling broader contexts, but rarely investigate whether the learned representations provide reliable evidence or introduce interference into classification decisions. To address this issue, we view hyperspectral image classification from the perspective of pixel-level evidence reliability modeling and propose PNEC-Mamba, a prototype-guided positive-negative evidence calibration framework. The framework progressively establishes semantic references, separates class-related evidence from interference, estimates pixel-level reliability, and performs selective calibration. First, a full-image state-space encoder extracts pixel representations, while dynamic class prototypes provide semantic references that evolve jointly with the feature space. Subsequently, positive and negative evidence is derived from pixel-prototype competition, explicitly separating discriminative cues that support classification from confusing signals associated with competing classes. Based on these evidence relationships, a multi-source uncertainty estimation strategy is introduced to assess pixel-level reliability, enabling stronger evidence calibration for uncertain regions. Finally, a full-resolution consistency refinement step is applied to recover local spatial details and improve boundary coherence in the final predictions. Extensive experiments on three benchmark datasets demonstrate that PNEC-Mamba achieves superior classification performance compared with state-of-the-art methods.

Problem

• hyperspectral image classification
• evidence reliability
• mixed pixels
• spectral ambiguity
• interference

Innovation

• evidence calibration
• prototype guidance
• state-space model
• hyperspectral image classification
• uncertainty estimation