USING L1-MAGIC FOR FEATURE ENHANCEMENT AND REDUCED REDUNDANCY IN HYPERSPECTRAL DATA

Author

• Ashley Alfred, University of Texas at ArlingtonFollow

Graduation Semester and Year

Summer 2026

Language

English

Document Type

Thesis

Degree Name

Doctor of Philosophy in Mathematics

Department

Mathematics

First Advisor

Jianzhong Su

Second Advisor

Hristo Kojouharov

Third Advisor

Ren-Cang Li

Fourth Advisor

Wei Jiang

Abstract

Hyperspectral imaging offers detailed spectral information, but achieving high spatial resolution typically requires large and expensive equipment. This study explores an alternative approach: enhancing low-quality hyperspectral bands using an L1-norm minimization technique known as L1-magic. The goal is to improve the utility of low-cost hardware by preserving discriminative features, promoting sparsity, and reducing spectral redundancy. We apply L1-magic to enhance low-quality bands and hypothesize that this method selectively amplifies key features while suppressing redundant information. Experimental results indicate that the enhanced bands approach the quality of high-resolution data, enabling robust feature extraction without reliance on high-end hyperspectral cameras.

Keywords

Spatial resolution, Spectral redundancy, Sparsity promotion, Band enhancement, Image quality, Compressive sensing, Optimization techniques, Mathematical modeling, Data dimensionality reduction

Disciplines

Agricultural Education | Agricultural Science | Numerical Analysis and Computation

License

This work is licensed under a Creative Commons Attribution 4.0 International License.

Recommended Citation

Alfred, Ashley, "USING L1-MAGIC FOR FEATURE ENHANCEMENT AND REDUCED REDUNDANCY IN HYPERSPECTRAL DATA" (2026). Mathematics Dissertations. 6.
https://mavmatrix.uta.edu/math_dissertations2/6