Evaluation of Encoding Schemas for Optimization of Bit-Level Run-Length Encoding Within Lossless Compression of Binary Images

摘要

The paper deals with the problematics of lossless data compression and is focused on the use of Run-Length Encoding (RLE) principle of lossless compression of binary images, especially binarized volume datasets that can be considered as the 3D binary images. Different alternative encoding schemas are proposed in the first part of this work for bit-level run-length encoding and capacities of their encoding words are discussed also with the focus on the encoding of long symbol runs. Test results of proposed encoding schemas using binarized volume datasets obtained by the medical imaging techniques including Computed Tomography (CT) and Magnetic Resonance Imaging (MRI) and the influence of the length of the encoding word to the compression rate of proposed encoding schemas are discussed in the second part of the paper.