Change detection methods for multi- and hypervariate data ideally identify differences in data acquired over the same area at different points in time. These differences may be due to noise or differences in (atmospheric etc.) conditions at the two acquisition time points. To prevent a change detection method from detecting uninteresting change due to noise or arbitrary spurious differences the application of regularisation also known as penalisation and other types of robustification of the change detection method are considered to be important. In this contribution an iterative extension to the multivariate alteration detection (MAD) transformation for change detection is described and applied. The MAD transformation is based on canonical correlation analysis (CCA) which is an established technique in multivariate statistics. The extension in an iterative scheme seeks to establish an increasingly better background of no-change against which to measure change. This is done by putting higher weights on observations of no-change in the calculation of the statistics for the CCA. Results from partly simulated multivariate data show an improved performance of the iterated scheme over the original MAD method. Also, a few comparisons with established methods for calculation of robust statistics for the CCA indicate that the scheme suggested here performs better. For this paper the new method will be applied to hyperspectral data also.
|Title of host publication||4th EARSeL Workshop on Imaging Spectroscopy|
|Publication status||Published - 2005|
Nielsen, A. A. (2005). An iterative extension to the MAD transformation for change detection in multi- and hyperspectral remote sensing data. In 4th EARSeL Workshop on Imaging Spectroscopy http://www2.imm.dtu.dk/pubdb/p.php?3529