Line Detection Methods for Spectrogram Images

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

Abstract

Accurate feature detection is key to higher level decisions regarding image content. Within the domain of spectrogram track detection and classification, the detection problem is compounded by low signal to noise ratios and high track appearance variation. Evaluation of standard feature detection methods present in the literature is essential to determine their strengths and weaknesses in this domain. With this knowledge, improved detection strategies can be developed. This paper presents a comparison of line detectors and a novel linear feature detector able to detect tracks of varying gradients. It is shown that the Equal Error Rates of existing methods are high, highlighting the need for research into novel detectors. Preliminary results obtained with a limited implementation of the novel method are presented which demonstrate an improvement over those evaluated.
Original languageEnglish
Title of host publicationComputer Recognition Systems
EditorsMarek Kurzynski, Michal Wozniak
PublisherSpringer
Pages127-134
Volume3
ISBN (Electronic)9783540939054
ISBN (Print)9783540939047
DOIs
Publication statusPublished - 1 May 2009
Event6th International Conference on Computer Recognition Systems - Jelenia Góra, Poland
Duration: 25 May 200928 May 2009

Publication series

NameAdvances in Intelligent and Soft Computing
PublisherSpringer
Volume57/2009
ISSN (Print)1615-3871
ISSN (Electronic)1860-0794

Conference

Conference6th International Conference on Computer Recognition Systems
Country/TerritoryPoland
CityJelenia Góra
Period25/05/0928/05/09

Bibliographical note

© 2009 Springer. This is an author produced version of the published paper. Uploaded in accordance with the publisher's self-archiving policy.

Cite this