Automatic point matching between SAR and optical images, match-point quality characterisation and error detection

We match multitemporal optical and SAR images of various spatial resolutions and spectral properties. The aims of such matching could be various, like image co-registration. Here, the SAR images are orthorectified and a related DSM exists. Thus, through matching we can automatically determine GCPs for optical images, which are needed for an accurate optical image georeferencing. However, this type of matching is extremely erroneous, so we developed novel methods for matching quality characterization and error detection. Very variable test datasets were used and the results are very promising, with poorer performance for low spatial resolution images.