Růžička, J. Impact of GDAL JPEG 2000 Lossy Compression to Digital Elevation Model

When dealing with a large set of geodata it is always necessary to use compression algorithms to lower their size. There are several algorithms that can be used for aerial (satellite) photos. The algorithms with a best compression ratio are nowadays based on a wavelet technique. Although digital elevation models in a GRID form are a bit different to aerial (satellite) photos the wavelet algorithms can be used for a compression of DEM. The compression can be loss or loss less. The paper describes results of tests that has been provided with GDAL software to produce high ratio compressed DEM in format JPEG 2000. The goal of this research was to determine how the loss compression impact quality of DEM. OpenJPEG library was used with several options to find how these options can change resulting compressed DEM. GRASS GIS was used to find out if there is a way how to pre-process DEM data before compression to reach better quality of the compressed DEM with keeping the same compression ratio.