Linear and kernel methods for multi- and hypervariate change detection

Allan Aasbjerg Nielsen, Morton John Canty

AbstractThe paper gives an overview of unsupervised, automatic change detection methods particularly the MAD method which is based on an iterated version of canonical correlation analysis with subsequent post-processing by means of linear and kernel versions of principal component analysis, maximum autocorrelation factor analysis, and minimum noise fraction analysis.
TypeConference paper [With referee]
ConferenceSPIE Europe Remote Sensing Conference 7830
Year2010    Month September
NoteInvited contribution
Electronic version(s)[pdf]
BibTeX data [bibtex]
IMM Group(s)Geoinformatics