Exploiting Color for Edge Extraction and Line Segment Multiview Matchingpdfauthor

This report investigates into the use of color information for edge extraction and straight line segment matching between two and more views. For edge extraction in color images different methods proposed in the literature are evaluated and compared, paying special attention to significance and completeness of the obtained edge-map. To find related line segment pairs in different views, we apply an oddman-out scheme: starting with all geometrically possible pairs in two views we first rule out pairs, for which the chromatic information provided by the regions flanking the line segments is incompatible. To further restrict the number of pairs we compute a chromatic similarity measure based on cross-correlation in the color bands. Both steps result in a significant reduction of candidate pairs, yet no correct pairs get lost. To reduce the number of wrong matches further, multiple views are integrated using the trifocal and the derived chromatic constraints. The presented method produces a highly reliable set of 3D line segments. A main application of our technique is for automatic 3D building reconstruction from high resolution aerial images.