Classification using Enhanced Spectral and Spatial Transformer with Grasshopper Optimization
Received: 6 November 2024 | Revised: 4 December 2024, 11 December 2024, and 17 December 2024 | Accepted: 20 December 2024 | Online: 2 February 2025
Corresponding author: Pilligundla Niharika
Abstract
Hyperspectral image (HSI) classification plays a crucial role in remote sensing, allowing the identification of various land cover types. Traditionally, Convolutional Neural Networks (CNNs) have been widely used for this purpose. However, they often face challenges related to high training parameter requirements and limited capacity for feature extraction, affecting their overall effectiveness. To overcome these challenges, this study proposes a novel approach integrating the Enhanced Deep Spectral and Spatial Transformer (EDSST) with Grasshopper Optimization (GHO). EDSST leverages transformer architecture to perform advanced spectral and spatial feature extraction, effectively mitigating the limitations of CNNs. This method improves feature abstraction and classification performance by reducing the number of training parameters while implementing a self-focusing mechanism. This approach incorporates a Classification Head (CH) with an orthogonal softmax activation function to accurately classify hyperspectral images. The proposed method was rigorously evaluated using the Salinas dataset, a benchmark in HSI classification research. The results show substantial improvements over existing techniques, achieving an accuracy of 99.5472%, precision of 99.5574%, recall of 99.5267%, and an F score of 99.6145%. These findings not only demonstrate the effectiveness of the proposed method in HSI classification but also highlight its efficiency and robustness, offering a promising solution for future applications in remote sensing and environmental monitoring.
Keywords:
convolutional neural networks, enhanced deep spectral and spatial transformer, grasshopper optimization, classification head, hyperspectral imagesDownloads
References
S. E. Qian, "Hyperspectral Satellites, Evolution, and Development History," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 7032–7056, 2021.
S. L. Ustin and E. M. Middleton, "Current and near-term advances in Earth observation for ecological applications," Ecological Processes, vol. 10, no. 1, Jan. 2021, Art. no. 1.
P. Mishra, A. Karami, A. Nordon, D. N. Rutledge, and J. M. Roger, "Automatic de-noising of close-range hyperspectral images with a wavelength-specific shearlet-based image noise reduction method," Sensors and Actuators B: Chemical, vol. 281, pp. 1034–1044, Feb. 2019.
R. Pillay, J. Y. Hardeberg, and S. George, "Hyperspectral imaging of art: Acquisition and calibration workflows," Journal of the American Institute for Conservation, vol. 58, no. 1–2, pp. 3–15, Apr. 2019.
M. Zhu, L. Jiao, F. Liu, S. Yang, and J. Wang, "Residual Spectral–Spatial Attention Network for Hyperspectral Image Classification," IEEE Transactions on Geoscience and Remote Sensing, vol. 59, no. 1, pp. 449–462, Jan. 2021.
S. Xu, M. Wang, X. Shi, Q. Yu, and Z. Zhang, "Integrating hyperspectral imaging with machine learning techniques for the high-resolution mapping of soil nitrogen fractions in soil profiles," Science of The Total Environment, vol. 754, Feb. 2021, Art. no. 142135.
F. Xie and Y. Xu, "An efficient regularized K-nearest neighbor structural twin support vector machine," Applied Intelligence, vol. 49, no. 12, pp. 4258–4275, Dec. 2019.
S. Swain, A. Banerjee, M. Bandyopadhyay, and S. C. Satapathy, "Dimensionality Reduction and Classification in Hyperspectral Images Using Deep Learning," in Machine Learning Approaches for Urban Computing, M. Bandyopadhyay, M. Rout, and S. Chandra Satapathy, Eds. Singapore: Springer, 2021, pp. 113–140.
D. Hong et al., "SpectralFormer: Rethinking Hyperspectral Image Classification With Transformers," IEEE Transactions on Geoscience and Remote Sensing, vol. 60, pp. 1–15, 2022.
M. E. Paoletti, J. M. Haut, N. S. Pereira, J. Plaza, and A. Plaza, "Ghostnet for Hyperspectral Image Classification," IEEE Transactions on Geoscience and Remote Sensing, vol. 59, no. 12, pp. 10378–10393, Sep. 2021.
H. Chen, F. Miao, Y. Chen, Y. Xiong, and T. Chen, "A Hyperspectral Image Classification Method Using Multifeature Vectors and Optimized KELM," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 2781–2795, 2021.
H. Sun, X. Zheng, and X. Lu, "A Supervised Segmentation Network for Hyperspectral Image Classification," IEEE Transactions on Image Processing, vol. 30, pp. 2810–2825, 2021.
S. Ghaderizadeh, D. Abbasi-Moghadam, A. Sharifi, N. Zhao, and A. Tariq, "Hyperspectral Image Classification Using a Hybrid 3D-2D Convolutional Neural Networks," IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 14, pp. 7570–7588, 2021.
X. Zheng, H. Sun, X. Lu, and W. Xie, "Rotation-Invariant Attention Network for Hyperspectral Image Classification," IEEE Transactions on Image Processing, vol. 31, pp. 4251–4265, 2022.
Y. Meraihi, A. B. Gabis, S. Mirjalili, and A. Ramdane-Cherif, "Grasshopper Optimization Algorithm: Theory, Variants, and Applications," IEEE Access, vol. 9, pp. 50001–50024, 2021.
"AVIRIS - Airborne Visible / Infrared Imaging Spectrometer - Data." https://aviris.jpl.nasa.gov/data/get_aviris_data.html.
Downloads
How to Cite
License
Copyright (c) 2025 P. Niharika, Shanker Chandre
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
