Principle component analysis based computing in image recognition

Charles Z. Liu*, Manolya Kavakli-Thorne

*Corresponding author for this work

Research output: Chapter in Book/Published conference outputChapter

Abstract

This chapter mainly focuses on Principle Component Analysis (PCA) based method for image computing. The underlying mechanism of PCA and several significant factors involved in subspace training are theoretically discussed including principle components' energy, residuals assessment, and decomposition computation. The typical extensions, including Probabilistic PCA (PPCA), Kernel PCA (KPCA), multidimensional PCA and Robust PCA (RPCA), have been critically analysed in this chapter.

Original languageEnglish
Title of host publicationImage Recognition
Subtitle of host publication Progress, Trends and Challenges
PublisherNova Science Publishers Inc
Pages31-66
Number of pages36
ISBN (Electronic)9781536172591
ISBN (Print)9781536172584
Publication statusPublished - 12 May 2020

Bibliographical note

Publisher Copyright:
© 2020 by Nova Science Publishers, Inc. All rights reserved.

Keywords

  • Eigenspace
  • Extended PCA
  • Image recognition
  • Karhunen-Loeve transform
  • Principle component analysis
  • Subspace training

Fingerprint

Dive into the research topics of 'Principle component analysis based computing in image recognition'. Together they form a unique fingerprint.

Cite this