• Home
  • Search
  • HyperSMamba: A Lightweight Mamba for Efficient Hyperspectral Image Classification
  • Cite Icon3
  • https://doi.org/10.3390/rs17122008Copy DOI Icon

HyperSMamba: A Lightweight Mamba for Efficient Hyperspectral Image Classification

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Deep learning has recently achieved remarkable progress in hyperspectral image (HSI) classification. Among these advancements, the Transformer-based models have gained considerable attention due to their ability to establish long-range dependencies. However, the quadratic computational complexity of the self-attention mechanism limits its application in hyperspectral image classification (HSIC). Recently, the Mamba architecture has shown outstanding performance in 1D sequence modeling tasks owing to its lightweight linear sequence operations and efficient parallel scanning capabilities. Nevertheless, its application in HSI classification still faces challenges. Most existing Mamba-based approaches adopt various selective scanning strategies for HSI serialization, ensuring the adjacency of scanning sequences to enhance spatial continuity. However, these methods lead to substantially increased computational overhead. To overcome these challenges, this study proposes the Hyperspectral Spatial Mamba (HyperSMamba) model for HSIC, aiming to reduce computational complexity while improving classification performance. The suggested framework consists of the following key components: (1) a Multi-Scale Spatial Mamba (MS-Mamba) encoder, which refines the state-space model (SSM) computation by incorporating a Multi-Scale State Fusion Module (MSFM) after the state transition equations of original SSMs. This module aggregates adjacent state representations to reinforce spatial dependencies among local features; (2) our proposed Adaptive Fusion Attention Module (AFAttention) to dynamically fuse bidirectional Mamba outputs for optimizing feature representation. Experiments were performed on three HSI datasets, and the findings demonstrate that HyperSMamba attains overall accuracy of 94.86%, 97.72%, and 97.38% on the Indian Pines, Pavia University, and Salinas datasets, while maintaining low computational complexity. These results confirm the model’s effectiveness and potential for practical application in HSIC tasks.

Similar Papers
  • Research Article
  • Citations1

Multi-Scale Hybrid Spectral Network for Feature Learning and Hyperspectral Image Classification

  • Jul 13, 2023
  • International Journal on Recent and Innovation Trends in Computing and Communication
  • Easala Ravi Kondal +1
  • Research Article
  • Citations43

Hyperspectral image classification based on discriminative locality preserving broad learning system

  • Jul 29, 2020
  • Knowledge-Based Systems
  • Yonghe Chu +6
  • Research Article
  • Citations1

3DVT: Hyperspectral Image Classification Using 3D Dilated Convolution and Mean Transformer

  • Feb 11, 2025
  • Photonics
  • Xinling Su +1
  • Conference Article
  • Citations1

Hyperspectral Image Classification Using 3D Attention Mechanism in Collaboration with Transformer

  • Sep 22, 2023
  • Yubing Wang +2
  • Research Article
  • Citations31

State space models meet transformers for hyperspectral image classification

  • Aug 22, 2024
  • Signal Processing
  • Xuefei Shi +6
  • Conference Article
  • Citations5

A Deep Learning approach to Hyperspectral Image Classification using an improved Hybrid 3D-2D Convolutional Neural Network

  • Sep 02, 2020
  • Dimitra Koumoutsou +1
  • Research Article
  • Citations1

Hyperspectral Image Classification Algorithm for Forest Analysis Based on a Group-Sensitive Selective Perceptual Transformer

  • Oct 19, 2024
  • Applied Sciences
  • Shaoliang Shi +3
  • Research Article
  • Citations84

BS2T: Bottleneck Spatial–Spectral Transformer for Hyperspectral Image Classification

  • Jan 01, 2022
  • IEEE Transactions on Geoscience and Remote Sensing
  • Ruoxi Song +4
  • Research Article
  • Citations8

HyperGCN – a multi-layer multi-exit graph neural network to enhance hyperspectral image classification

  • Jul 05, 2024
  • International Journal of Remote Sensing
  • Haseena Rahmath P +3
  • PDF
  • Research Article
  • Citations38

Residual Group Channel and Space Attention Network for Hyperspectral Image Classification

  • Jun 24, 2020
  • Remote Sensing
  • Peida Wu +3
  • Research Article
  • Citations6

Segment Anything Model-Based Hyperspectral Image Classification for Small Samples

  • Apr 10, 2025
  • Remote Sensing
  • Kaifeng Ma +6
  • PDF
  • Research Article
  • Citations14

Small Sample Hyperspectral Image Classification Method Based on Dual-Channel Spectral Enhancement Network

  • Aug 13, 2022
  • Electronics
  • Songwei Pei +2
  • PDF
  • Research Article
  • Citations17

Improving Hyperspectral Image Classification with Compact Multi-Branch Deep Learning

  • Jun 07, 2024
  • Remote Sensing
  • Md Rashedul Islam +3
  • Research Article
  • Citations3

Fractal-Based Ensemble Classification System for Hyperspectral Images

  • Jan 01, 2023
  • IEEE Geoscience and Remote Sensing Letters
  • Behnam Asghari Beirami +2
  • PDF
  • Research Article
  • Citations62

FusionNet: A Convolution–Transformer Fusion Network for Hyperspectral Image Classification

  • Aug 19, 2022
  • Remote Sensing
  • Liming Yang +6
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.