Skip to content
Open access

FEAM-Swin: A Lightweight Frequency Aware Swin Transformer for Efficient Hyperspectral Image Classification

2026 · Computer Modeling in Engineering & Sciences · Vol 148, pp. 1-10 · 0 citations · 87 references

TL;DR

A novel Frequency-Enhanced Attention Modulator (FEAM), which captures local spectral–spatial frequency variations via gradient-based energy estimation and performs adaptive channel modulation with minimal computational overhead, and is embedded within Swin Transformer blocks to enhance discriminative feature learning while preserving the efficiency of window-based self-attention.

Abstract

: Hyperspectral image (HSI) classification requires models that can effectively capture long-range contextual dependencies while preserving fine-grained spectral–spatial variations under strict computational constraints. Recent transformer-based approaches, particularly Swin Transformers, have shown strong performance by leveraging localized self-attention; however, their reliance on generic attention mechanisms often overlooks frequency-sensitive information that is critical for discriminating spectrally similar materials. Moreover, existing frequency-aware designs typically introduce heavy parameterization or explicit spectral transforms, limiting their efficiency and practical deployment. In this paper, we propose FEAM-Swin, a lightweight frequency-aware Swin Transformer designed for efficient HSI classification. The proposed model introduces a novel Frequency-Enhanced Attention Modulator (FEAM), which captures local spectral–spatial frequency variations via gradient-based energy estimation and performs adaptive channel modulation with minimal computational overhead. Unlike conventional frequency modelling approaches, FEAM avoids explicit Fourier or wavelet transforms, enabling seamless integration into hierarchical transformer architectures. FEAM is embedded within Swin Transformer blocks to enhance discriminative feature learning while preserving the efficiency of window-based self-attention. Extensive experiments performed on four benchmark hyperspectral datasets widely used in the research community, including Indian Pines, University of Pavia, Salinas and KSC, demonstrate that FEAM-Swin consistently outperforms state-of-the-art CNN-and transformer-based methods. Specifically, FEAM-Swin achieves overall accuracies of 98.17%, 99.87%, 99.97%, and 98.41%, respectively, yielding consistent improvements over competing approaches while requiring fewer parameters and lower computational complexity. These results validate the effectiveness of lightweight frequency-aware modulation and establish FEAM-Swin as a practical and robust solution for real-world HSI classification.

Read PDF

Similar papers

Open access 2026

DDFGFormer: Dual-Decoupled Frequency-Guided Transformer for Hyperspectral Image Classification

Hyperspectral image (HSI) classification requires jointly capturing subtle spectral differences and complex spatial structures. Recent studies have introduced frequency-domain modeling to complement spatial–spectral representation learning. However, existing frequency-based methods still face some limitations. First, m...

Jin-Liang An, Deng-Ru Han, Long-Long Dai et al. · 0 citations
Open access 2026

SSLANet: A Frequency–Spatial Visual System for Efficient Hyperspectral Image Classification

Hyperspectral image classification requires a compact representation that can jointly capture discriminative spectral cues, local spatial structures, and efficient inference behavior under limited annotations. Existing CNN-based methods provide effective local pattern extraction but are less capable of modeling long-ra...

Xiao-Fei Yang, Rui-Ying Zhang, Wei-Jia Cao et al. · 0 citations
2026

FA3DHeat: A Frequency-Aware 3-D Heat Conduction Network for Hyperspectral Image Classification

Hyperspectral image (HSI) classification requires effective spectral–spatial representation learning, since spectral signatures and spatial structures are highly coupled. Although convolutional and attention-based methods have achieved promising performance, efficient modeling of spectral–spatial information in the int...

Dong Shi, Zhi-Kang Xiang, Yu Shen et al. · 0 citations
Open access 2026

Frequency-Aware Bias Calibration Attention Network for Hyperspectral Image Classification

Recently, self-attention mechanisms have significantly advanced hyperspectral image classification (HSIC) by enabling effective modeling of long-range spatial–spectral dependencies. However, existing methods often suffer from blurred land-cover boundaries and insufficient discriminability between spectrally similar cla...

Zhen-Qiu Shu, Nan-Bo Liu, Ke-Xin Zeng et al. · 0 citations
Open access 2026

RWKVHSI: Dual-Stream Recurrent Spectral–Spatial Evidence Fusion for Efficient Hyperspectral Image Classification

Hyperspectral image (HSI) classification requires reconciling highly correlated spectral responses with spatially heterogeneous structures while controlling computational cost. Convolutional networks provide strong local inductive biases but capture long-range context only indirectly, whereas self-attention improves gl...

Xiao-Fei Yang, Xiao-Lin Luo, Zi-Hang Luo et al. · 0 citations
Open access 2026

SMGNet: A Lightweight Spectral-Mamba-Guided Network for Hyperspectral Image Classification

SMGNet is proposed, a lightweight spectral-Mamba-guided network that establishes a “spectral leads, spatial follows” paradigm that employs a state space model (Mamba) to efficiently capture global long-range spectral dependencies with linear complexity, while a lightweight Res-CNN preserves intact spatial topology.

Xiao-Yang Meng, Long-Bao Wang, Kai-Jun Jia et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.