Uzaktan Algılamada Çoklu-Set Kanonik Korelasyon Analizi ile Değişiklik Tespiti

Abstract

In this study, a new change detection approach is proposed for remote sensing domain. It is shown that the proposed method can be employed to expose land cover alteration in remotely sensed imagery.The proposed method is based on multi-set canonical correlation analysis, which is an outstanding procedure for revealing correlated information between multi-set variables. Multi-set canonical correlation analysis makes it possible to obtain maximally correlated variables from linear combinations of input multi-set variables. Analysis can be carried out in sets consisting of fixed spectral variables in different times or various spectral variables belonging to fixed acquisition times.European Space Agency has made Culture-MERIS product freely available for the scientific community. The proposed method and related change detection methods are applied on the Culture-MERIS for comparison and performance assessment.