Work

A Prompt-Guided Transformer Network With Redundancy-Aware Endmember Bundles Considering Spectral Variability for Hyperspectral Unmixing

Year: 2026

Type: article

Abstract: Spectral variability (SV) in complex environments brings a dual challenge to accurate hyperspectral unmixing (HU): large intraclass and small interclass differences. Most deep learning (DL) frameworks treat features uniformly and fail to distinguish highly similar endmembers, which easily confuses abundance estimation. To simultaneously tackle this dual challenge, integrating endmember bundle priors that effectively accommodate intraclass variations into deep networks has emerged as a promising solution. However, current methods face bottlenecks in both the extraction and utilization of endmember bundles. To this end, we propose a novel prompt-guided transformer network (PGT-Net). It accommodates intraclass variations while effectively amplifying subtle interclass differences. First, addressing the redundancy and noise introduced by existing extraction methods that blindly pursue quantity, a density-based redundancy-aware endmember bundle extraction (DR-EBE) algorithm is proposed. It utilizes density-guided clustering and a spatial–spectral screening strategy to extract high-quality, low-redundancy endmember bundle priors. Second, our innovatively designed prompt-guided endmember transformer (PGET) converts these priors into endmember-specific prompt vectors. It performs targeted activation of specific features to capture and amplify subtle interclass differences. Finally, a spatial complementary integration module (SCIM) is designed. It utilizes spatial complementarity to adaptively refine feature boundaries and performs cross-channel interactions to further break information isolation. Extensive experiments on multiple hyperspectral datasets demonstrate the effectiveness and competitive performance of PGT-Net compared with existing unmixing algorithms. The code is publicly available athttps://github.com/UPCGIT/PGT-Net.

Source: IEEE Transactions on Geoscience and Remote Sensing

DOI: 10.1109/tgrs.2026.3715348