A Quadtree-based Progressive Lossless Compression Technique for Volumetric Data Sets
Gregor Klajnsek, Borut Zalik, Franc Novak* and Gregor Papa* Faculty of Electrical Engineering and Computer Science University of Maribor Smetanova 17, 2000 Maribor, Slovenia *Jozef Stefan Institute Jamova 39, 1000 Ljubljana, Slovenia
An efficient technique for progressive lossless compression of volumetric data is described. It is based on the quadtree data structure and exploits the expected similarities between the neighboring slices. The proposed approach has better compression ratio than the octree method. A small amount of memory is required since only two slices need to be located in the memory at a time, which makes it suitable for hardware implementation. A built-in compression in volumetric scanners could considerably reduce the transfer and storage requirements for volumetric medical data.