Regularized Difference Criterion for Computing Discriminants for Dimensionality Reduction

Alex J. Aved, Erik P. Blasch, Jing Peng

Research output: Contribution to journalArticlepeer-review

13 Scopus citations


Hyperspectral data classification has shown potential in many applications. However, a large number of spectral bands cause overfitting. Methods for reducing spectral bands, e.g., linear discriminant analysis, require matrix inversion. We propose a semidefinite programming for linear discriminants regularized difference (SLRD) criterion approach that does not require matrix inversion. The paper establishes a classification error bound and provides experimental results with ten methods over six hyperspectral datasets demonstrating the efficacy of the proposed SLRD technique.

Original languageEnglish
Article number7906579
Pages (from-to)2372-2384
Number of pages13
JournalIEEE Transactions on Aerospace and Electronic Systems
Issue number5
StatePublished - Oct 2017


  • Classification
  • dimensionality reduction
  • feature selection
  • hyperspectral imaging


Dive into the research topics of 'Regularized Difference Criterion for Computing Discriminants for Dimensionality Reduction'. Together they form a unique fingerprint.

Cite this