At its most basic level, many common compression algorithms rely on coding digital words of variable length, or VLC, Huffman type, designed in 1952, or any of its derivatives, as canonical. This article explains "Leveling," a VLC that is more efficient than Huffman encoding and has two variants: "Leveled Reordering" for minimal redundancy, such as text, and "Segmented Leveling" for moderate and high redundancy, such as image processing. In 1995, Javier Joglar created levelling, which uses the concepts of "meaning" and "ordering" of the VLC codes generated to obtain the maximum "compression ratio" of any non-adaptive VLC.
Author(s) Details
Javier Joglar
Alcubilla
Freelance Researcher, Madrid, Spain.
No comments:
Post a Comment