Laser Radar (LADAR) achieves more applications on aerial aided-navigation in mountainous areas for its good
performance. But plain areas encounter terrain elevation's slow variation and occasional unavailability of Digital Feature
Analysis Database (DFAD), which as necessary reference. Looking for replaceable map source and extracting common
characters for matching, are the fundamental circles of imaging LADAR aided navigation research. In this paper aerial
high-resolution remote sensing (RS) images are applied as substitute for DFAD, and the edge factor is chosen out by
synthetically analyzing RS images' and imaging LADAR point cloud'scharacters. Then edge extraction algorithm based
on multi-scale wavelet is explored to reflect their common features, and weighted Hausdorff distance method is applied
to match for positioning. At last the high-resolution RS images and imaging LADAR data of the same area are assumed
for simulation experiment, which testifies the validity of the methods proposed above.
|