mHC-HSI: Clustering-Guided Hyper-Connection Mamba for Hyperspectral Image Classification

2026-03-03

AI Summary

Existing hyperspectral image classification methods struggle to simultaneously achieve effective spatial-spectral feature fusion and model interpretability. This work proposes a clustering-guided manifold-constrained hyperconnected (mHC)-Mamba module that decomposes the input into spectrally coherent groups with physical meaning via soft clustering and constructs a multi-stream parallel state-space model. The residual matrices in this architecture are interpreted as soft clustering membership maps, thereby endowing the model with explicit interpretability. Evaluated on multiple hyperspectral benchmark datasets, the proposed method significantly outperforms current state-of-the-art approaches, achieving higher classification accuracy while enhancing the transparency and physical plausibility of learned features.

Technology Category

• Computer Vision: Interpretability, Explainability, and Transparency
• Machine Learning: Transparent, Interpretable, Explainable ML
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Recently, DeepSeek has invented the manifold-constrained hyper-connection (mHC) approach which has demonstrated significant improvements over the traditional residual connection in deep learning models \cite{xie2026mhc}. Nevertheless, this approach has not been tailor-designed for improving hyperspectral image (HSI) classification. This paper presents a clustering-guided mHC Mamba model (mHC-HSI) for enhanced HSI classification, with the following contributions. First, to improve spatial-spectral feature learning, we design a novel clustering-guided Mamba module, based on the mHC framework, that explicitly learns both spatial and spectral information in HSI. Second, to decompose the complex and heterogeneous HSI into smaller clusters, we design a new implementation of the residual matrix in mHC, which can be treated as soft cluster membership maps, leading to improved explainability of the mHC approach. Third, to leverage the physical spectral knowledge, we divide the spectral bands into physically-meaningful groups and use them as the "parallel streams" in mHC, leading to a physically-meaningful approach with enhanced interpretability. The proposed approach is tested on benchmark datasets in comparison with the state-of-the-art methods, and the results suggest that the proposed model not only improves the accuracy but also enhances the model explainability. Code is available here: https://github.com/GSIL-UCalgary/mHC_HyperSpectral

Problem

• hyperspectral image classification
• spatial-spectral feature learning
• model explainability
• heterogeneous data
• spectral prior knowledge

Innovation

• clustering-guided
• hyper-connection
• Mamba
• hyperspectral image classification
• spectral grouping