Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Optimal Hyperspectral Undersampling Strategy for Satellite Imaging

Vita V. Vlasova, Vladimir G. Kuzmin, Maria S. Varetsa, Natalia A. Ibragimova, Oleg Y. Rogov, Elena V. Lyapuntsova

TL;DR

This work tackles hyperspectral image classification under high dimensionality and limited labeled data by introducing Iterative Wavelet-based Gradient Sampling (IWGS), a band-selection method that uses gradient information in the wavelet domain to iteratively retain informative spectral bands. Coupled with a lightweight hybrid CNN-SSM backbone, IWGS achieves strong classification performance while reducing memory and compute demands, making it suitable for edge devices. The approach is validated on benchmark data (e.g., Indian Pines and Houston 2013), demonstrating superior accuracy and robustness to atmospheric noise and adversarial perturbations, with high OA, AA, and $\kappa$ metrics. Overall, the method advances practical, efficient hyperspectral analytics by aligning discriminative band selection with robust, scalable spectral-spatial modeling.

Abstract

Hyperspectral image (HSI) classification presents significant challenges due to the high dimensionality, spectral redundancy, and limited labeled data typically available in real-world applications. To address these issues and optimize classification performance, we propose a novel band selection strategy known as Iterative Wavelet-based Gradient Sampling (IWGS). This method incrementally selects the most informative spectral bands by analyzing gradients within the wavelet-transformed domain, enabling efficient and targeted dimensionality reduction. Unlike traditional selection methods, IWGS leverages the multi-resolution properties of wavelets to better capture subtle spectral variations relevant for classification. The iterative nature of the approach ensures that redundant or noisy bands are systematically excluded while maximizing the retention of discriminative features. We conduct comprehensive experiments on two widely-used benchmark HSI datasets: Houston 2013 and Indian Pines. Results demonstrate that IWGS consistently outperforms state-of-the-art band selection and classification techniques in terms of both accuracy and computational efficiency. These improvements make our method especially suitable for deployment in edge devices or other resource-constrained environments, where memory and processing power are limited. In particular, IWGS achieved an overall accuracy up to 97.8% on Indian Pines for selected classes, confirming its effectiveness and generalizability across different HSI scenarios.

Optimal Hyperspectral Undersampling Strategy for Satellite Imaging

TL;DR

This work tackles hyperspectral image classification under high dimensionality and limited labeled data by introducing Iterative Wavelet-based Gradient Sampling (IWGS), a band-selection method that uses gradient information in the wavelet domain to iteratively retain informative spectral bands. Coupled with a lightweight hybrid CNN-SSM backbone, IWGS achieves strong classification performance while reducing memory and compute demands, making it suitable for edge devices. The approach is validated on benchmark data (e.g., Indian Pines and Houston 2013), demonstrating superior accuracy and robustness to atmospheric noise and adversarial perturbations, with high OA, AA, and metrics. Overall, the method advances practical, efficient hyperspectral analytics by aligning discriminative band selection with robust, scalable spectral-spatial modeling.

Abstract

Hyperspectral image (HSI) classification presents significant challenges due to the high dimensionality, spectral redundancy, and limited labeled data typically available in real-world applications. To address these issues and optimize classification performance, we propose a novel band selection strategy known as Iterative Wavelet-based Gradient Sampling (IWGS). This method incrementally selects the most informative spectral bands by analyzing gradients within the wavelet-transformed domain, enabling efficient and targeted dimensionality reduction. Unlike traditional selection methods, IWGS leverages the multi-resolution properties of wavelets to better capture subtle spectral variations relevant for classification. The iterative nature of the approach ensures that redundant or noisy bands are systematically excluded while maximizing the retention of discriminative features. We conduct comprehensive experiments on two widely-used benchmark HSI datasets: Houston 2013 and Indian Pines. Results demonstrate that IWGS consistently outperforms state-of-the-art band selection and classification techniques in terms of both accuracy and computational efficiency. These improvements make our method especially suitable for deployment in edge devices or other resource-constrained environments, where memory and processing power are limited. In particular, IWGS achieved an overall accuracy up to 97.8% on Indian Pines for selected classes, confirming its effectiveness and generalizability across different HSI scenarios.
Paper Structure (17 sections, 3 equations, 1 figure, 3 tables, 1 algorithm)

This paper contains 17 sections, 3 equations, 1 figure, 3 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Deep Learning in Hyperspectral Image Analysis
  4. Transformers in Hyperspectral Image Classification
  5. Synthesis and Outlook
  6. Methodology
  7. Iterative Wavelet-based Gradient Sampling (IWGS)
  8. Adversarial Perturbations
  9. Noise Model
  10. Classification Objective.
  11. Adversarial PGD Perturbations.
  12. Robust Classification Objective.
  13. Experiments
  14. Experimental Setup
  15. Adversarial robustness.
  16. ...and 2 more sections

Figures (1)

  • Figure 1: Classification map generated by IGWS compared to ground truth on the Indian pines dataset. Despite operating under limited data, IGWS effectively preserves class boundaries and structure, demonstrating strong generalization and class balance. Coloring is artificial and added to guide the eye.