Novel gabor-PHOG features for object and scene image classification

Atreyee Sinha, Sugata Banerji, Chengjun Liu

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Scopus citations

Abstract

A new Gabor-PHOG (GPHOG) descriptor is first introduced in this paper for image feature extraction by concatenating the Pyramid of Histograms of Oriented Gradients (PHOG) of all the local Gabor filtered images. Next, a comparative assessment of the classification performance of the GPHOG descriptor is made in six different color spaces, namely the RGB, HSV, YCbCr, oRGB, DCS and YIQ color spaces, to propose the novel YIQ-GPHOG and the YCbCr-GPHOG feature vectors that perform well on different object and scene image categories. Third, a novel Fused Color GPHOG (FC-GPHOG) feature is presented by integrating the PCA features of the six color GPHOG descriptors for object and scene image classification, with applications to image search and retrieval. Finally, the Enhanced Fisher Model (EFM) is applied for discriminatory feature extraction and the nearest neighbor classification rule is used for image classification. The effectiveness of the proposed feature vectors for image classification is evaluated using two grand challenge datasets, namely the Caltech 256 dataset and the MIT Scene dataset.

Original languageEnglish (US)
Title of host publicationStructural, Syntactic, and Statistical Pattern Recognition - Joint IAPR International Workshop, SSPR and SPR 2012, Proceedings
Pages584-592
Number of pages9
DOIs
StatePublished - 2012
EventJoint IAPR International Workshops on Structural and Syntactic PatternRecognition, SSPR 2012 and Statistical Techniques in Pattern Recognition,SPR 2012 - Hiroshima, Japan
Duration: Nov 7 2012Nov 9 2012

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7626 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherJoint IAPR International Workshops on Structural and Syntactic PatternRecognition, SSPR 2012 and Statistical Techniques in Pattern Recognition,SPR 2012
Country/TerritoryJapan
CityHiroshima
Period11/7/1211/9/12

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Keywords

  • EFM
  • FC-GPHOG
  • Gabor-PHOG (GPHOG)
  • PCA
  • YCbCr-GPHOG
  • YIQ-GPHOG
  • color spaces
  • image search

Fingerprint

Dive into the research topics of 'Novel gabor-PHOG features for object and scene image classification'. Together they form a unique fingerprint.

Cite this