Content-based image retrieval using color, texture and shape features

Alexey Ponomarev, Hitesh S. Nalamwar, Ilya Babakov, Chandrakant S. Parkhi, Gaurav Buddhawar

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

7 Citations (Scopus)

Abstract

Nowadays CBIR is getting more and more attention from organizations and researchers due to advances in digital imaging techniques. A lot of interest is getting paid to search images from large databases, as it is not only difficult and time-consuming task but sometimes frustrating for the users. This paper proposes the CBIR system based on color, texture and shape features. The proposed method employs the use of DCT and DWT along with Hierarchical k-means algorithm for faster retrieval of images. The efficiency of the given method is demonstrated by the results in the paper.

Original languageEnglish
Title of host publicationHigh Technology
Subtitle of host publicationResearch and Applications, 2015
EditorsNikita Martyushev, Anna Bogdan
PublisherTrans Tech Publications Ltd
Pages872-876
Number of pages5
ISBN (Print)9783038357087
DOIs
Publication statusPublished - 2016
Event4th International Conference for Young Scientists High Technology: Research and Applications, HTRA 2015 - Tomsk, Russian Federation
Duration: 21 Apr 201524 Apr 2015

Publication series

NameKey Engineering Materials
Volume685
ISSN (Print)1013-9826
ISSN (Electronic)1662-9795

Conference

Conference4th International Conference for Young Scientists High Technology: Research and Applications, HTRA 2015
CountryRussian Federation
CityTomsk
Period21.4.1524.4.15

Keywords

  • (discrete cosine transform) DCT
  • (discrete wavelet transform) DWT
  • Content-based image retrieval (CBIR)
  • Hierarchical k-means
  • Range filtering
  • Text on co-occurrence matrix

ASJC Scopus subject areas

  • Materials Science(all)
  • Mechanics of Materials
  • Mechanical Engineering

Fingerprint Dive into the research topics of 'Content-based image retrieval using color, texture and shape features'. Together they form a unique fingerprint.

Cite this