Representation of images by the Delaunay triangulation
|Inserted by||Doc. Ing. Josef Kohout, Ph.D.|
|Date last modified||29.12.2013|
|Number of downloads||3|
The software takes a grey-scale raster image as input and transforms it into a vector image represented by the Delaunay triangulation using the chosen heuristics. Generally, there are two utilities included in the software: DOTr, which starts with a coarse representation that is refined by adding other pixels to get the representation of the requested quality, and TriImgCompress, which starts with a fine representation that is decimated by removing the least significant pixels to get the representation of the requested quality. Several methods within both utilities are available.
The methods implemented in this software, experiments and their results are described in the following papers:
Kohout J: Alternative Representation of Image Information. Technical Report, University of West Bohemia, 2009, (DCSE/TR-2009-11)
Kohout J, Kolingerová I, Janák T: Compression of images using Delaunay triangulation. In: Proceedings of the 9th IASTED International Conference on Vizualization, Imaging, and Image Processing, 2009.