Combined morphological-spectral unsupervised image segmentation

RJ O'Callaghan, DR Bull

Research output: Contribution to journalArticle (Academic Journal)peer-review

142 Citations (Scopus)


Probably the best unsupervised segmentation algorithm. Used as a basis for mutimodality segmentation and fusion in work funded by DIF DTC. Evaluated in trial airborne platform by QinetiQ, producing superior results to all other state of the art algorithms tried. Selected for pull through into Phase II of DIF DTC in the Multidimensional Fusion Cluster Project lead by General Dynamics. Acknowledged to be the foundation stone of our successful groundbreaking work in region based image and video fusion. Used as basis for work in the successful Link Autoarch project on archiving and metadata extraction for Wildlife archives.
Translated title of the contributionCombined morphological-spectral unsupervised image segmentation
Original languageEnglish
Article numberIssue 1
Pages (from-to)49 - 62
Number of pages14
JournalIEEE Transactions on Image Processing
Volume14 (1)
Publication statusPublished - Jan 2005

Bibliographical note

Publisher: Institute of Electrical and Electronics Engineers (IEEE)


Dive into the research topics of 'Combined morphological-spectral unsupervised image segmentation'. Together they form a unique fingerprint.

Cite this