Local-global joint decision based clustering for airport recognition

Bingxin Qu, Yanning Zhang, Tao Yang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Airport recognition plays an important role in many computer vision tasks. This paper proposes a novel method for airport recognition including the generation of clustering threshold adaptively and automatically based on local-global joint decision. The novelties of the approach include: (1) a new adaptive clustering framework we called local-global joint decision based clustering is designed instead of a fixed threshold. (2) Within the special framework, we propose a statistical probability model based on a new local feature as the preliminary decision strategy which provide local information. (3) Another part of the framework we called reconfirmation strategy contains an innovative relative similarity model that uses global information to estimate the confidence coefficient of the result from preliminary decision. Extensive experimental results demonstrate the feasibility of our approach. In addition, our method outperforms the related algorithms in terms of recognition accuracy under a variety of situations.

Original languageEnglish
Title of host publicationIntelligence Science and Big Data Engineering - 4th International Conference, IScIDE 2013, Revised Selected Papers
PublisherSpringer Verlag
Pages94-102
Number of pages9
ISBN (Print)9783642420566
DOIs
StatePublished - 2013
Event4th International Conference on Intelligence Science and Big Data Engineering, IScIDE 2013 - Beijing, China
Duration: 31 Jul 20132 Aug 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8261 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference4th International Conference on Intelligence Science and Big Data Engineering, IScIDE 2013
Country/TerritoryChina
CityBeijing
Period31/07/132/08/13

Keywords

  • adaptive clustering
  • Airport recognition
  • local-global joint decision

Fingerprint

Dive into the research topics of 'Local-global joint decision based clustering for airport recognition'. Together they form a unique fingerprint.

Cite this