Sports image classification through Bayesian classifier

Youna Jung, Eenjun Hwang, Wonil Kim

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

8 Citations (Scopus)

Abstract

Most documents on the web today contain image as well as text data. So far, classification of images has been dependent on the annotation. For the more efficient and accurate classification, not only textual data but also image contents should be considered. In this paper, we propose a novel method for classifying specific image data; sports images. The proposed method is based on the Bayesian framework and employs four important color features to exploit the properties of sports images.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
EditorsRicardo Conejo, Jose-Luis Perez-de-la-Cruz, Maite Urretavizcaya
PublisherSpringer Verlag
Pages546-555
Number of pages10
ISBN (Print)3540222189, 9783540222187
DOIs
Publication statusPublished - 2004
Externally publishedYes
Event10th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2003 and 5th Conference on Technology Transfer, TTIA 2003 - San Sebastian, Spain
Duration: 2003 Nov 122003 Nov 14

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3040
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference10th Conference of the Spanish Association for Artificial Intelligence, CAEPIA 2003 and 5th Conference on Technology Transfer, TTIA 2003
Country/TerritorySpain
CitySan Sebastian
Period03/11/1203/11/14

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'Sports image classification through Bayesian classifier'. Together they form a unique fingerprint.

Cite this